With the progression of Neural Architecture Search (NAS), it has increasingly caught the attention of researchers from various domains. Up to now, researchers have developed numerous Demand-Driven NAS (DD-NAS) methods to alleviate difficulties of constructing deep neural network architectures across various practical applications. However, there lacks a review that summarizes the technical characteristics of NAS methods from a demand-driven perspective. Thus, in this paper, we provide a comprehensive survey of the latest advancements in DD-NAS. Specifically, first of all, we outline the fundamental process and core components of DD-NAS. Then, we summarize the unique characteristics of DD-NAS methods used in four main domains. Following this, we take representative DD-NAS methods based on electroencephalogram as an example to conduct empirical researches. Finally, we explore potential directions for the future development of DD-NAS.

The automatic generation of brain CT reports has gained widespread attention, given its potential to assist radiologists in diagnosing cranial diseases. However, brain CT scans involve extensive medical entities, such as diverse anatomy regions and lesions, exhibiting highly inconsistent spatial patterns in 3D volumetric space. This leads to biased learning of medical entities in existing methods, resulting in repetitiveness and inaccuracy in generated reports. To this end, we propose a Medical Entity-balanced Prompting Network (MEPNet), which harnesses the large language model (LLM) to fairly interpret various entities for accurate brain CT report generation. By introducing the visual embedding and the learning status of medical entities as enriched clues, our method prompts the LLM to balance the learning of diverse entities, thereby enhancing reports with comprehensive findings. First, to extract visual embedding of entities, we propose Knowledge-driven Joint Attention to explore and distill entity patterns using both explicit and implicit medical knowledge. Then, a Learning Status Scorer is designed to evaluate the learning of entity visual embeddings, resulting in unique learning status for individual entities. Finally, these entity visual embeddings and status are elaborately integrated into multi-modal prompts, to guide the text generation of LLM. This process allows LLM to self-adapt the learning process for biased-fitted entities, thereby covering detailed findings in generated reports. We conduct experiments on two brain CT report generation benchmarks, showing the effectiveness in clinical accuracy and text coherence.

Traffic prediction provides vital support for urban traffic management and has received extensive research interest. By virtue of the ability to effectively learn spatial and temporal dependencies from a global view, Transformers have achieved superior performance in long-term traffic prediction. However, existing methods usually underrate the complex spatio-temporal entanglement in long-range sequences. Compared with purely temporal entanglement, spatio-temporal data emphasizes the entangled dynamics under the restrictions of traffic networks, which brings additional difficulties. Moreover, the computational costs of spatio-temporal Transformers scale quadratically as the sequence length grows, limiting their applications on long-range and large-scale scenarios. To address these problems, we propose a decomposed spatio-temporal Mamba (DST-Mamba) for traffic prediction. We aim to apply temporal decomposition to the entangled sequences and obtain the seasonal and trend parts. Shifting from the temporal view to the spatial view, we leverage Mamba, a state space model with near-linear complexity, to capture seasonal variations in a node-centric manner. Meanwhile, multi-scale trend information is extracted and aggregated by simple linear layers. Such combination equips DST-Mamba with superior capability to model long-range spatio-temporal dependencies while remaining efficient compared with Transformers. Experimental results across five real-world datasets demonstrate that DST-Mamba can capture both local fluctuations and global trends within traffic patterns, achieving state-of-the-art performance with favorable efficiency.

Deep models are increasingly used to analyze brain graphs for the diagnosis and understanding of brain diseases. However, due to the multi-site data aggregation and individual differences, brain graph datasets exhibit widespread distribution shifts, which impair the model’s generalization ability to the test set, thereby limiting the performance of existing methods. To address these issues, we propose a Causally Invariance-aware Augmentation for brain Graph Contrastive Learning, called CIA-GCL. This method first generates a brain graph by extracting node features based on the topological structure. Then, a learnable brain invariant subgraph is identified based on a causal decoupling approach to capture the maximum label-related invariant information with invariant learning. Around this invariant subgraph, we design a novel invariance-aware augmentation strategy to generate meaningful augmented samples for graph contrast learning. Finally, the extracted invariant subgraph is utilized for brain disease classification, effectively mitigating distribution shifts while also identifying critical local graph structures, enhancing the model’s interpretability. Experiments on three real-world brain disease datasets demonstrate that our method achieves state-of-the-art performance, effectively generalizes to multi-site brain datasets, and provides certain interpretability.

Inferring causal protein signaling networks from human immune system cellular data is an important approach to reveal underlying tissue signaling biology and dysfunction in diseased cells. In recent years, reinforcement learning (RL) methods have shown excellent performance in the field of causal protein signaling network inference. However, the complexity of RL models and the need for manual hyperparameter tuning can hinder performance. In this paper, we propose a actor-critic RL model via artificial bee colony (ABC) neural architecture search, called ABCNAS-RL. Specifically, the entire method is divided into two phases: ABC neural architecture search and actor-critic RL search. In phase one, we represent each bee as a set of hyperparameter, utilizing the ABC algorithm searching for optimal hyperparameters of the actor-critic RL model on the training set. In phase two, we use the actor-critic RL model to infer the causal protein signaling network on the test set. The actor network consists of an encoder-decoder architecture, composed of a transformer and a bidirectional gated recurrent unit (BiGRU) with an integrated attention mechanism. The critic network consists of a fully connected neural network that estimates the output state of the actor network. By maximizing cumulative rewards, we ultimately derive the causal protein signaling network. Extensive experimental results on simulated and real datasets verify that ABCNAS-RL outperforms the comparison methods and has superior performance.

Estimating brain effective connectivity (EC) from functional magnetic resonance imaging (fMRI) data can aid in comprehending the neural mechanisms underlying human behavior and cognition, providing a foundation for disease diagnosis. However, current spatiotemporal attention modules handle temporal and spatial attention separately, extracting temporal and spatial features either sequentially or in parallel. These approach overlooks the inherent spatiotemporal correlations present in real world fMRI data. Additionally, the presence of noise in fMRI data further limits the performance of existing methods. In this paper, we propose a novel brain effective connectivity estimation method based on Fourier spatiotemporal attention (FSTA-EC), which combines Fourier attention and spatiotemporal attention to simultaneously capture inter-series (spatial) dynamics and intra-series (temporal) dependencies from high-noise fMRI data. Specifically, Fourier attention is designed to convert the high-noise fMRI data to frequency domain, and map the denoised fMRI data back to physical domain, and spatiotemporal attention is crafted to simultaneously learn spatiotemporal dynamics. Furthermore, through a series of proofs, we demonstrate that incorporating learnable filters into fast Fourier transform and inverse fast Fourier transform processes is mathematically equivalent to performing cyclic convolution. The experimental results on simulated and real-resting-state fMRI datasets demonstrate that the proposed method exhibits superior performance when compared to state-of-the-art methods. The code is available at https://github.com/XiongWenXww/FSTA.

Pre-trained Large language models (LLMs) have shown impressive advancements in functional magnetic resonance imaging (fMRI) analysis and causal discovery. Considering the unique nature of the causal discovery field, which focuses on extracting causal graphs from observed data, research on LLMs in this field is still at an early exploratory stage. As a subfield of causal discovery, effective connectivity (EC) has received even less attention, and LLM-based approaches in EC remain unexplored. Existing LLM-based approaches for causal discovery typically rely on iterative querying to assess the causal influence between variable pairs, without any model adaptation or fine-tuning, making them ill-suited for handling the cross-modal gap and complex causal structures. To this end, we propose BrainEC-LLM, the first method to fine-tune LLMs for estimating brain EC from fMRI data. Specifically, multiscale decomposition mixing module decomposes fMRI time series data into short-term and long-term multiscale trends, then mixing them in bottom-up (fine to coarse) and top-down (coarse to fine) manner to extract multiscale temporal variations. And cross attention is applied with pre-trained word embeddings to ensure consistency between the fMRI input and pre-trained natural language. The experimental results on simulated and real resting-state fMRI datasets demonstrate that BrainEC-LLM can achieve superior performance when compared to state-of-the-art baselines.

Spatio-temporal neural networks have been successfully applied to weather forecasting tasks recently. The key notion is to learn spatio-temporal features concurrently from spatial and temporal dependencies. Existing methods are mainly based on local smoothness assumptions where the features are learned by accumulating information in local spatio-temporal regions. However, the weather conditions in a certain spatio-temporal region are usually influenced by global meteorological changes and long-range historical weather conditions. Therefore, these methods that ignore the large-scale spatio-temporal effects can hardly learn effective features. In this paper, we propose a novel spatio-temporal Transformer network in weather forecasting to address the above challenges. The main idea is to leverage the Transformer architecture to carefully capture the multi-scale spatial and long-range temporal information in weather data. First, we propose to combine the global and local position encodings based on absolute geographic locations and relative geodesic distances and insert them into the spatial Transformer to extract the multi-scale spatial information in meteorological graphs. Then, we further capture the long-range temporal dependencies by a temporal Transformer where the attention mechanism is used to improve the representation ability and scalability of the models. Extensive experiments over real weather datasets demonstrate the effectiveness of our framework.

Multi-head attention (MA), which allows the model to jointly attend to crucial information from diverse representation subspaces through its heads, has yielded remarkable achievement in image captioning. However, there is no explicit mechanism to ensure MA attends to appropriate positions in diverse subspaces, resulting in overfocused attention for each head and redundancy between heads. In this paper, we propose a novel Intra- and Inter-Head Orthogonal Attention (I2OA) to efficiently improve MA in image captioning by introducing a concise orthogonal regularization to heads. Specifically, Intra-Head Orthogonal Attention enhances the attention learning of MA by introducing orthogonal constraint to each head, which decentralizes the object-centric attention to more comprehensive content-aware attention. Inter-Head Orthogonal Attention reduces the heads redundancy by applying orthogonal constraint between heads, which enlarges the diversity of representation subspaces and improves the representation ability for MA. Moreover, the proposed I2OA is flexible to combine with various multi-head attention based image captioning methods and improve the performances without increasing model complexity and parameters. Experiments on the MS COCO dataset demonstrate the effectiveness of the proposed model.

In recent years, semi-supervised learning (SSL) for functional brain network (FBN) classification has gained considerable attention due to its potential to leverage large amounts of unlabeled data from multisite. However, existing SSL methods often struggle to address the distributional differences across different sites, which limits their ability to extract discriminative features from the unlabeled data, thus hindering classification performance. To overcome this challenge, we propose a novel consistency semi-supervised FBN classification framework with prototypical-adversarial learning, termed CSBNC-PAL. Specifically, we first design a contrastive consistency module (CCM) that utilizes contrastive learning to more effectively exploit unlabeled data and learn preliminary feature representations. Then, we introduce a prototype alignment module (PAM) that computes site-aware prototypes through weighted feature clustering to guide inter-site feature alignment, and achieve inter-site equilibrium feature representations. Finally, we develop an adversarial alignment module (AAM) that employs site-discriminative adversarial training based on a gradient reversal layer to guide intra-site feature alignment, and learn site-invariant features. The three modules above are optimized collectively in an end-to-end manner, ensuring effective learning from both labeled and unlabeled data while alleviating the distribution differences of multisite data. Experiments on the ABIDE I, ABIDE II, and ADHD-200 datasets demonstrate that the CSBNC-PAL outperforms many state-of-the-art SSL methods in FBN classification.

Due to the limitations of collection conditions and costs, public brain network datasets generally combine data from multiple sites. However, the difference among multi-source data collected from multiple sites always affects classification performance. To overcome the problem, we propose a triplet-based deep hashing incremental learning (Tri-DHIL) method for brain network classification, which learns data streams from each site incrementally rather than from collections of multiple sites. Specifically, the Tri-DHIL method is divided into three phases. In the site queue generation phase, we rank the sites based on their sample quantity and label information. In the triplet-based deep hashing learning phase, we first cluster samples of the site using diagnostic labels and take the clustering center as the anchor point. Then, we choose two samples to form a triplet with the anchor point, one from the same cluster as the anchor point and the other from a different cluster. The construction of triplets can not only enrich the input data of the model but also facilitate the maintenance of similarity relationships in the process of model learning. Finally, we input the triplets into the deep hashing learning model for feature extraction and hash mapping. In the incremental learning phase, we adjust the model parameters by accumulating the triplet-based losses of the previous site, which can prevent the model from forgetting the previously learned features after learning the features of the new site. Experimental results on ABIDE I, ABIDE II, and ADHD-200 demonstrate that the Tri-DHIL method exhibits competitive classification performance.

In recent years, the discovery of brain effective connectivity (EC) networks through computational analysis of functional magnetic resonance imaging (fMRI) data has gained prominence in neuroscience and neuroimaging. However, owing to the influence of diverse factors during data collection and processing, fMRI data typically exhibits high noise and limited sample characteristics, consequently leading to suboptimal performance of current methods. In this paper, we propose a novel brain effective connectivity discovery method based on meta-reinforcement learning, called MetaRLEC. The method mainly consists of three modules: actor, critic, and meta-critic. MetaRLEC first employs an encoder-decoder framework: the encoder utilizing a Transformer, converts noisy fMRI data into a state embedding; the decoder employing bidirectional LSTM, discovers brain region dependencies from the state and generates actions (EC networks). Then a critic network evaluates these actions, incentivizing the actor to learn higher-reward actions amidst the high-noise setting. Finally, a meta-critic framework facilitates online learning of historical state-action pairs, integrating an action-value neural network and supplementary training losses to enhance the model's adaptability to small-sample fMRI data. We conduct comprehensive experiments on both simulated and real-world data to demonstrate the efficacy of our proposed method.

Weather forecasting has become a popular research topic recently, which mainly benefits from the development of spatio-temporal neural networks to effectively extract useful patterns from weather data. Generally, the weather changes in the meteorological system are governed by physical principles. However, it is challenging for spatio-temporal methods to capture the physical knowledge of meteorological dynamics. To address this problem, we propose in this paper a spatio-temporal Transformer network with physical knowledge distillation (PKD-STTN) for weather forecasting. First, the teacher network is implemented by a differential equation network that models weather changes by the potential energy in the atmosphere to reveal the physical mechanism of atmospheric movements. Second, the student network uses a spatio-temporal Transformer that concurrently utilizes three attention modules to comprehensively capture the semantic spatial correlation, geographical spatial correlation, and temporal correlation from weather data. Finally, the physical knowledge of the teacher network is transferred to the student network by inserting a distillation position encoding into the Transformer. Notice that the output of the teacher network is distilled to the position encoding rather than the output of the student network, which can largely utilize physical knowledge without influencing the feature extraction process of Transformers. Experiments on benchmark datasets show that the proposed method can effectively utilize physical principles of weather changes and has obvious performance advantages compared with several strong baselines.

Dynamic effective connectivity (DEC) is the accumulation of effective connectivity in the time dimension, which can describe the continuous neural activities in the brain. Recently, learning DEC from functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) data has attracted the attention of neuroinformatics researchers. However, the current methods fail to consider the gap between the fMRI and EEG modality, which can not precisely learn the DEC network from multimodal data. In this paper, we propose a multimodal causal adversarial network for DEC learning, named MCAN. The MCAN contains two modules: multimodal causal generator and multimodal causal discriminator. First, MCAN employs a multimodal causal generator with an attention-guided layer to produce a posterior signal and output a set of DEC networks. Then, the proposed method uses a multimodal causal discriminator to unsupervised calculate the joint gradient, which directs the update of the whole network. The experimental results on simulated data sets show that MCAN is superior to other state-of-the-art methods in learning the network structure of DEC and can effectively estimate the brain states. The experimental results on real data sets show that MCAN can better reveal abnormal patterns of brain activity and has good application potential in brain network analysis.

Learning brain effective connectivity networks (ECN) from functional magnetic resonance imaging (fMRI) data has gained much attention in recent years. With the successful applications of deep learning in numerous fields, several brain ECN learning methods based on deep learning have been reported in the literature. However, current methods ignore the deep temporal features of fMRI data and fail to fully employ the spatial topological relationship between brain regions. In this article, we propose a novel method for learning brain ECN based on spatiotemporal graph convolutional models (STGCM), named STGCMEC, in which we first adopt the temporal convolutional network to extract the deep temporal features of fMRI data and utilize the graph convolutional network to update the spatial features of each brain region by aggregating information from neighborhoods, which makes the features of brain regions more discriminative. Then, based on such features of brain regions, we design a joint loss function to guide STGCMEC to learn the brain ECN, which includes a task prediction loss and a graph regularization loss. The experimental results on a simulated dataset and a real Alzheimer's disease neuroimaging initiative (ADNI) dataset show that the proposed STGCMEC is able to better learn brain ECN compared with some state-of-the-art methods.

Multimodal multiobjective optimization problems (MMOPs) have attracted wide attention in recent years. This kind of problem is very challenging since they need to locate different Pareto-optimal solution sets (PSs) that correspond to the same Pareto front. To resolve it, this article proposes a novel multimodal multiobjective differential evolution (DE) algorithm with species conservation, which develops a new way of locating different PSs. Specifically, the proposed algorithm adopts species conservation to determine different PSs in known areas, while it uses a variant of DE as the basic optimizer to explore new areas. There are three operators in species conservation: 1) species division; 2) seed determination; and 3) seed conservation. Species division mainly partitions the joint population of parents and children into various species in the decision space for retaining different PSs. Seed determination selects superior solutions from each species as seeds that need to be kept in the next generation. Seed conservation is to ensure that all species seeds are retained in the new generation by substituting no promising solutions with them, thereby guarantee not missing some known areas that may contain different PSs. Besides, the DE variant is utilized to produce diverse solutions to find new areas in the decision space where PSs may exist. The comparative experiments with ten state-of-the-art algorithms have been performed on the CEC 2019 MMOPs test set and two real-world problems. The experimental results have verified that the proposed algorithm has a competitive performance for MMOPs.

Recently, the functional brain network (FBN) classification methods based on deep neural networks (DNNs) have around a lot of scientific interest. However, these DNN architectures are manually designed by human experts through trial-and-error testing, which not only requires rich parameter tuning experience and large labor costs, but also a fixed manual architecture cannot consistently guarantee good performance across different data distributions and scenarios. To solve this problem, we propose an adaptive particle swarm architecture search method based on multi-level convolutions, which can automatically design suitable DNN architectures for various FBN classification tasks. Specifically, to effectively extract multi-level features at FBN, we construct three multi-level convolution units to form candidate architectures. These units can extract edge-level, node-level, and graph-level features respectively. The parameters of these units will be searched using the particle swarm-based NAS framework. Additionally, to alleviate the difficulty of searching in a vast search space, we propose a novel adaptive updating strategy. This strategy adaptively locks specific elements of the particle vector based on historical information and the search epochs, which can effectively search within a subset of the vast search space. We conduct systematic experiments on ABIDE I, ABIDE II, and ADHD-200 datasets with different atlases. The experimental results demonstrate that our method achieves competitive accuracies of 74.71%, 73.03%, and 74.39% on the CC200 atlas, and 71.42%, 73.91%, and 69.96% on the AAL atlas respectively.

The automatic Brain CT reports generation can improve the efficiency and accuracy of diagnosing cranial diseases.However, current methods are limited by 1) coarse-grained supervision: the training data in image-text format lacks detailed supervision for recognizing subtle abnormalities, and 2) coupled cross-modal alignment: visual-textual alignment may be inevitably coupled in a coarse-grained manner, resulting in tangled feature representation for report generation.In this paper, we propose a novel Pathological Graph-driven Cross-modal Alignment (PGCA) model for accurate and robust Brain CT report generation.Our approach effectively decouples the cross-modal alignment by constructing a Pathological Graph to learn finegrained visual cues and align them with textual words.This graph comprises heterogeneous nodes representing essential pathological attributes (i.e., tissue and lesion) connected by intra-and inter-attribute edges with prior domain knowledge.Through carefully designed graph embedding and updating modules, our model refines the visual features of subtle tissues and lesions and aligns them with textual words using contrastive learning.Extensive experimental results confirm the viability of our method.We believe that our PGCA model holds the potential to greatly enhance the automatic generation of Brain CT reports and ultimately contribute to improved cranial disease diagnosis.

Recently, clinical phenotypic semantic information has begun to play an important role in some brain network classification methods based on deep learning. However, most current methods only consider the phenotypic semantic information of individual brain networks but ignore the potential phenotypic characteristics among group brain networks. To address this problem, we present a deep hashing mutual learning (DHML)-based brain network classification method. Specifically, we first design a separable CNN-based deep hashing learning to extract individual topological features of brain networks and map them into hash codes. Secondly, we construct a group brain network relationship graph based on the similarity of phenotypic semantic information, in which each node is a brain network, and the properties of the nodes are the individual features extracted in the previous step. Then, we adopt a GCN-based deep hashing learning to extract the group topological features of the brain network and map them to hash codes. Finally, the two deep hashing learning models perform mutual learning by measuring the distribution differences between the hash codes to achieve the interaction of individual and group features. The experimental results on the three commonly used brain atlases (AAL Atlas, Dosenbach160 Atlas, and CC200 Atlas) of the ABIDE I dataset show that our proposed DHML method achieves optimal classification performance compared with some state-of-the-art methods.

Spatiotemporal graph neural networks (GNNs) have been used successfully in traffic prediction in recent years, primarily owing to their ability to model complex spatiotemporal dependencies within irregular traffic networks. However, the feature extraction processes in these methods are limited in their exploration of the inner properties of traffic data. Specifically, graph and temporal convolutions are local operations and can hardly utilize information from wider ranges, which may affect the long-term prediction performance of such methods. Furthermore, deep spatiotemporal GNNs easily suffer from poor generalization owing to overfitting. To address these problems, this study presents a novel traffic prediction method that integrates self-supervised learning and self-distillation into spatiotemporal GNNs. First, a self-supervised learning module is used to explore the knowledge from the input data. An auxiliary task based on temporal continuity is designed to capture the contextual information in traffic data. Second, a self-distillation framework is developed as an implicit regularization approach that transfers knowledge from the model itself. The combination of self-supervision and self-distillation further mines the knowledge from the data and the model, and the generalization ability and stability of the prediction model can be improved. The proposed model achieved superior or competitive results compared with several strong baselines on six traffic prediction datasets. In particular, the maximum performance improvement ratios for the six datasets were 3.0% (MAE), 5.2% (RMSE), and 3.8% (MAPE). These results demonstrate the effectiveness of the proposed method.

Human brain effective connectivity characterizes the causal effects of neural activities among different brain regions. Studies of brain effective connectivity networks (ECNs) for different populations contribute significantly to the understanding of the pathological mechanism associated with neuropsychiatric diseases and facilitate finding new brain network imaging markers for the early diagnosis and evaluation for the treatment of cerebral diseases. A deeper understanding of brain ECNs also greatly promotes brain-inspired artificial intelligence (AI) research in the context of brain-like neural networks and machine learning. Thus, how to picture and grasp deeper features of brain ECNs from functional magnetic resonance imaging (fMRI) data is currently an important and active research area of the human brain connectome. In this survey, we first show some typical applications and analyze existing challenging problems in learning brain ECNs from fMRI data. Second, we give a taxonomy of ECN learning methods from the perspective of computational science and describe some representative methods in each category. Third, we summarize commonly used evaluation metrics and conduct a performance comparison of several typical algorithms both on simulated and real datasets. Finally, we present the prospects and references for researchers engaged in learning ECNs.

In recent years, using functional magnetic resonance imaging (fMRI) data to infer brain effective connectivity (EC) between different brain regions is an important advanced study in neuroinformatics. However, current methods always perform not well due to the high noise of neuroimaging data. In this paper, we propose an effective connectivity learning method with deep reinforcement learning, called EC-DRL, aiming to more accurately identify the brain effective connectivity from fMRI data. The proposed method is based on the actor-critic algorithm framework, using the encoder-decoder model as the actor network. More specifically, the encoder adopts the Transformer model structure, and the decoder uses a bidirectional long-short-term memory network with an attention mechanism. A large number of experimental results on simulated fMRI data and real-world fMRI data show that EC-DRL can better infer effective connectivity compared to the state-of-the-art methods.

Medical report automatic generation has gained increasing interest recently as a way to help radiologists write reports more efficiently. However, this image-to-text task is rather challenging due to the typical data biases: 1) Normal physiological structures dominate the images, with only tiny abnormalities; 2) Normal descriptions accordingly dominate the reports. Existing methods have attempted to solve these problems, but they neglect to exploit useful information from similar historical cases. In this paper, we propose a novel Cross-modal Contrastive Attention (CMCA) model to capture both visual and semantic information from similar cases, with mainly two modules: a Visual Contrastive Attention Module for refining the unique abnormal regions compared to the retrieved case images; a Cross-modal Attention Module for matching the positive semantic information from the case reports. Extensive experiments on two widely-used benchmarks, IU X-Ray and MIMIC-CXR, demonstrate that the proposed model outperforms the state-of-the-art methods on almost all metrics. Further analyses also validate that our proposed model is able to improve the reports with more accurate abnormal findings and richer descriptions.

Classification of dynamic functional connectivity (DFC) is becoming a promising approach for diagnosing various neurodegenerative diseases. However, the existing methods generally face the problem of overfitting. To solve it, this paper proposes a convolutional neural network with three sparse strategies named SCNN to classify DFC. Firstly, an element-wise filter is designed to impose sparse constraints on the DFC matrix by replacing the redundant elements with zeroes, where the DFC matrix is specially constructed to quantify the spatial and temporal variation of DFC. Secondly, a 1×1 convolutional filter is adopted to reduce the dimensionality of the sparse DFC matrix, and remove meaningless features resulted from zero elements in the subsequent convolution process. Finally, an extra sparse optimization classifier is employed to optimize the parameters of the above two filters, which can effectively improve the ability of SCNN to extract discriminative features. Experimental results on multiple resting-state fMRI datasets demonstrate that the proposed model provides a better classification performance of DFC compared with several state-of-the-art methods, and can identify the abnormal brain functional connectivity.

As a novel deep learning method, deep forest has achieved excellent classification performance on many small-scale datasets, thus providing a new opportunity to accurately classify brain networks (BNs) on limited fMRI data. Though there are a few explorations about classifying BNs using deep forest, they only adopt sliding windows to extract adjacent features of BNs and fail to use prior knowledge to strengthen the features more relevant to brain diseases. In this paper, we propose a deep forest framework with multi-channel message passing and neighborhood aggregation mechanisms (DF-MCMPNA) to extract and aggregate long-range multi-channel topological features. Firstly, we use the three intrinsic connectivity networks (ICNs) and the whole-brain to form four feature extraction channels. Secondly, we present a multi-channel message passing mechanism and a channel-shared neighborhood aggregation mechanism to recursively extract long-range multi-channel topological features, where the first mechanism can learn local topological features in each channel and the second mechanism can fuse multi-channel topological features. Finally, the extracted features are fed into the casForst to perform further feature learning and classification. Experimental results on ABIDE I, ABIDE II, and ADHD-200 datasets show that the DF-MCMPNA outperforms several state-of-the-art methods on classification performance and accurately identifies abnormal brain regions.

Brain network classification using resting-state functional magnetic resonance imaging (rs-fMRI) is an effective analytical method for diagnosing brain diseases. In recent years, brain network classification methods based on deep learning have attracted increasing attention. However, these methods only consider the spatial topological characteristics of the brain network but ignore its proximity relationships in semantic space. To overcome this problem, we propose a novel brain network classification method based on deep graph hashing learning named BNC-DGHL. Specifically, we first extract the deep features of the brain network and then learn a graph hash function based on clinical phenotype labels and the similarity of diagnostic labels. Secondly, we use the learned graph hash function to convert deep features into hash codes, which can maintain the original semantic spatial relationships. Finally, we calculate the distance between hash codes to obtain the predicted category of the brain network. Experimental results on ABIDE I, ABIDE II, and ADHD-200 datasets demonstrate that our method achieves better classification performance of brain diseases compared with some state-of-the-art methods, and the abnormal functional connectivities between brain regions identified may serve as biomarkers associated with related brain diseases.

Inferring brain-effective connectivity networks from neuroimaging data has become a very hot topic in neuroinformatics and bioinformatics. In recent years, the search methods based on Bayesian network score have been greatly developed and become an emerging method for inferring effective connectivity. However, the previous score functions ignore the temporal information from functional magnetic resonance imaging (fMRI) series data and may not be able to determine all orientations in some cases. In this article, we propose a novel score function for inferring effective connectivity from fMRI data based on the conditional entropy and transfer entropy (TE) between brain regions. The new score employs the TE to capture the temporal information and can effectively infer connection directions between brain regions. Experimental results on both simulated and real-world data demonstrate the efficacy of our proposed score function.

Report-writing for Brain Computed Tomography (CT) imaging is a routine procedure for diagnosing cerebrovascular diseases, while it is time-consuming and tedious for radiologists especially in highly populated areas. Automatic report generation has the potential to alleviate radiologists’ workload and reduce the diagnose error. Currently, the development of image captioning and medical image processing has driven great achievements in medical report generation. However, there is no report generation study for the Brain CT imaging and this task faces the following challenges: First, Brain CT lesions are disperse in 3-D space, with more morphological instability. Second, the Brain CT reports are long paragraphs with similar medical term. These challenges increase the difficulty o f lesions recognition and report generation for Brain CT imaging. To cope with these challenges, we propose a weakly guided hierarchical encoder-decoder network for lesions learning and Brain CT report generation. Specifically, we propose a weakly guided attention model (WGAM) in encoder to capture the most important areas and scans gradually under the weak guidance of possible lesions areas. In addition, we propose a keywords-driven interactive recurrent network (KIRN) in decoder to generate paragraphs under the weak guidance of possible lesions keywords. Experiments on our Brain CT dataset demonstrate the effectiveness of the proposed method.

Learning brain effective connectivity networks (ECNs) by means of deep learning methods from functional magnetic resonance imaging (fMRI) data is a novel study hot in neuroinformatics in recent years. However, current methods need manually tune and set a lot of model hyper-parameters. Once the parameter setting is unreasonable, it will seriously restrict the performance of algorithms. In this paper, we propose a novel method for learning ECNs based on controllable variational autoencoder (CVAE), named as CVAEEC. It can automatically tune model parameters and learn brain effective connectivity. In detail, the proposed method first adopts an encoder network to obtain the latent variables from the fMRI data of brain regions. And then, based on the latent variables, it utilizes a decoder network to obtain the generated fMRI data of brain regions. Once the generated fMRI data is highly similar to real fMRI data by iteratively training, CVAEEC algorithm can output an optimal brain ECN. The experimental results on a real dataset show that the proposed CVAEEC is able to better learn brain ECN compared to some state-of-the-art methods.

The functional connectivity provides new insights into the mechanisms of the human brain at network-level, which has been proved to be an effective biomarker for brain disease classification. Recently, machine learning methods have played an important role in functional connectivity classification, among which convolutional neural network (CNN) based methods become a new hot topic since they can extract topological features in the brain network. However, the conventional CNN-based methods haven't taken sparse connectivity patterns (SCPs) of the human brain into consideration, which may lead to redundancy of the topological features, and limit their performance and generalization. To solve it, we propose a novel CNN-based model with graphical Lasso (CNNGLasso) to extract sparse topological features for brain disease classification. First, we develop a novel graphical Lasso model for revealing the SCPs at group-level. Then, the SCPs are used to guide the topological feature extraction. Finally, the obtained sparse topological features are used to classify the patients from normal controls. The experiment results on the ABIDE dataset demonstrate that the CNNGLasso outperforms the others on various performances. Besides, the abnormal brain regions derived from the trained model are consistent with the previous investigations, which further proves the application prospect of the CNNGLasso.

Estimating effective connectivity from functional magnetic resonance imaging (fMRI) time series data has become a very hot topic in neuroinformatics and brain informatics. However, it is hard for the current methods to accurately estimate the effective connectivity due to the high noise and small sample size of fMRI data. In this paper, we propose a novel framework for estimating effective connectivity based on recurrent generative adversarial networks, called EC-RGAN. The proposed framework employs the generator that consists of a set of effective connectivity generators based on recurrent neural networks to generate the fMRI time series of each brain region, and uses the discriminator to distinguish between the joint distributions of the real and generated fMRI time series. When the model is well-trained and generated fMRI data is similar to real fMRI data, EC-RGAN outputs the effective connectivity by means of the causal parameters of the effective connectivity generators. Experimental results on both simulated and real-world fMRI time series data demonstrate the efficacy of our proposed framework.

Inferring effective connectivity between different brain regions from functional magnetic resonance imaging (fMRI) data is an important advanced study in neuroinformatics in recent years. However, current methods have limited usage in effective connectivity studies due to the high noise and small sample size of fMRI data. In this paper, we propose a novel framework for inferring effective connectivity based on generative adversarial networks (GAN), named as EC-GAN. The proposed framework EC-GAN infers effective connectivity via an adversarial process, in which we simultaneously train two models: a generator and a discriminator. The generator consists of a set of effective connectivity generators based on structural equation models which can generate the fMRI time series of each brain region via effective connectivity. Meanwhile, the discriminator is employed to distinguish between the joint distributions of the real and generated fMRI time series. Experimental results on simulated data show that EC-GAN can better infer effective connectivity compared to other state-of-the-art methods. The real-world experiments indicate that EC-GAN can provide a new and reliable perspective analyzing the effective connectivity of fMRI data.

Visual attention has been successfully applied in image captioning to selectively incorporate the most relevant areas to the language generation procedure. However, the attention in current image captioning methods is only guided by the hidden state of language model, e.g. LSTM (Long-Short Term Memory), indirectly and implicitly, and thus the attended areas are weakly relevant at different time steps. Besides the spatial relationship of attention areas, the temporal relationship in attention is crucial for image captioning according to the attention transmission mechanism of human vision. In this paper, we propose a new spatio-temporal memory attention (STMA) model to learn the spatio-temporal relationship in attention for image captioning. The STMA introduces the memory mechanism to the attention model through a tailored LSTM, where the new cell is used to memorize and propagate the attention information, and the output gate is used to generate attention weights. The attention in STMA transmits with memory adaptively and dependently, which builds strong temporal connections of attentions and learns the spatio-temporal relationship of attended areas simultaneously. Besides, the proposed STMA is flexible to combine with attention-based image captioning frameworks. Experiments on MS COCO dataset demonstrate the superiority of the proposed STMA model in exploring the spatio-temporal relationship in attention and improving the current attention-based image captioning.

Learning brain effective connectivity (EC) networks from functional magnetic resonance imaging (fMRI) data has become a new hot topic in the neuroinformatics field. However, how to accurately and efficiently learn brain EC networks is still a challenging problem. In this paper, we propose a new algorithm to learn the brain EC network structure using ant colony optimization (ACO) algorithm combining with voxel activation information, named as VACOEC. First, VACOEC uses the voxel activation information to measure the independence between each pair of brain regions and effectively restricts the space of candidate solutions, which makes many unnecessary searches of ants be avoided. Then, by combining the global score increase of a solution with the voxel activation information, a new heuristic function is designed to guide the process of ACO to search for the optimal solution. The experimental results on simulated datasets show that the proposed method can accurately and efficiently identify the directions of the brain EC networks. Moreover, the experimental results on real-world data show that patients with Alzheimers disease (AD) exhibit decreased effective connectivity not only in the intra-network within the default mode network (DMN) and salience network (SN), but also in the inter-network between DMN and SN, compared with normal control (NC) subjects. The experimental results demonstrate that VACOEC is promising for practical applications in the neuroimaging studies of geriatric subjects and neurological patients.

The dynamic nature of the brain functional connectivity (FC) is well accepted in recent years. However, most of the current FC classification methods are based on the static estimation of FC. In this paper, we propose a novel convolutional neural network with an element-wise filter for classifying dynamic functional connectivity (DFC-CNN). First, a DFC matrix is estimated to quantify the DFC. Then, taking the DFC matrix as input, the DFC-CNN model employs one-dimensional convolutional kernels to extract the high-level features of DFC. Moreover, an element-wise filter is specially designed for the DFC matrix, which further improves the classification performance. The experimental results on the autism brain imaging data exchange I (ABIDE I) indicate that the proposed model can distinguish subject groups more accurately, and also can be used to identify the abnormal brain regions.

Deep neural networks have been successfully applied to the classification of brain networks. However, the high-dimensional and small-scale properties of the brain network data limit their extensive applications. To solve this problem, this paper proposes a new deep forest framework with cross-shaped window scanning mechanism (DF-CWSM) to extract topological features for the classification of brain networks. The cross-shaped window scanning mechanism is designed to extract the node-level and the edge-level features respectively that have meaningful interpretations in terms of corresponding network topologies. Based on the classification framework, we firstly implement the feature transformation of brain networks by the multi-level topological feature extraction. Then a cascade forest structure is used to learn the hierarchical features layer by layer. And the results of the last level of cascade forests are integrated to make the final classification. We evaluated the proposed framework on the ABIDE I data set. Experimental results show that our proposed framework can not only achieve competitive classification performance but also accurately identify the abnormal brain regions associated with ASD.

Estimating brain effective connectivity (EC) from neuroimaging data has recently received wide interest and become a new topic in the neuroinformatics field. Currently, dynamic Bayesian networks (DBN) have been successfully applied to estimating EC from functional magnetic resonance imaging (fMRI) time-series data as they can capture the temporal characteristics of connectivity among brain regions. However, DBN methods assume that activations of brain regions are stationary and follow a Markovian condition, which are strong assumptions that may not be valid in many cases. In this paper, we introduce a novel method to estimate brain effective connectivity networks from fMRI data using non-stationary dynamic Bayesian networks, named as EC-nsDBN. EC-nsDBN can not only capture the non-stationary temporal information from fMRI time-series data but also estimate how interactions among brain regions change dynamically over the fMRI experiments. Systematic experiments on simulated data show that EC-nsDBN has better direction identification ability compared with other state-of-the-art algorithms, and can accurately capture the temporal characteristics of connectivity. Experiments on real-world data sets are also provided to support our analysis.

BACKGROUND: Biomedical semantic indexing is important for information retrieval and many other research fields in bioinformatics. It annotates biomedical citations with Medical Subject Headings. In face of unbalanced category distribution in the training data, sampling methods are difficult to apply for semantic indexing task. RESULTS: In this paper, we present a novel deep serial multi-task learning model. The primary task treats the biomedical semantic indexing as a multi-label text classification issue that considers the relations of the labels. The auxiliary task is a regression task that predicts the MeSH number of the citation and provides hints for the network to make it converge faster. The experimental results on the BioASQ-Task5A open dataset show that our model outperforms the state-of-the-art solution "MTI", proposed by the US National Library of Medicine. Further, it not only achieves the highest precision among all the solutions in BioASQ-Task5A but also has faster convergence speed compared with some naive deep learning methods. CONCLUSIONS: Rather than parallel in an ordinary multi-task structure, the tasks in our model are serial and tightly coupled. It can achieve satisfied performance without any handcrafted feature.

Biomedical semantic indexing refers to annotating biomedical citations with Medical Subject Headings, which is crucial for texting mining, information retrieval and other researches in the field of bioinformatics. The traditional methods ignore the relations among labels and need complicated feature engineering. In this paper, we present a novel model with a deep serial multi-task learning structure, in which the semantic word embedding and bidirectional Gated Recurrent Unit are integrated in a multi-task learning paradigm. It differs from an ordinary multi-task structure in that the tasks in our model are serial and tightly coupled rather than parallel. The dataset of the 2017 BioASQ-Task5A is used to evaluate the performance. Without any handcrafted feature, our model outperforms MTI, the state-of-the-art solution proposed by the US National Library of Medicine. It also achieves the highest precision among all the solutions in 2017 BioASQ-Task5A, and converges faster than some naive deep learning methods.

As protein-protein interactions always change with time, environments and different stages of cell cycle, the clustering analysis on static protein-protein interaction (PPI) networks can not reflect this dynamics property and is far from satisfactory. To solve it, this paper proposes a method based on time-sequenced association and Ant Colony Clustering for identifying Protein Complexes in Dynamic PPI networks (called ACC-DPC). ACC-DPC first splits a PPI network into a series of dynamics subnetworks under different time points by integrating gene expression data, and then makes the clustering analysis on each subnetwork using the ant colony clustering method. For each subnetwork, ACC-DPC begins with constructing initial protein clusters by introducing the time-sequenced association characteristic of protein complexes between two adjacent time points, and later uses the picking up and dropping down operators of ant colony clustering to accomplish the clustering process of other proteins. The experimental results on two PPI datasets demonstrate that ACC-DPC has competitive performances in identifying protein complexes of dynamic PPI networks compared with several algorithms.

Identifying brain effective connectivity networks from functional magnetic resonance imaging (fMRI) data is an important advanced subject in neuroinformatics in recent years, where the learning method based on bayesian networks (BN) has become a new hot topic in the field. This paper proposes a new method to learn the brain effective connectivity network structure by combining ant colony optimization (ACO) with BN method, named as ACOEC. In the proposed algorithm, a brain effective connectivity network is first mapped onto an ant, and then the ant colony optimization by simulating real ants looking for food is employed to construct network structures and finally an ant with the highest score is obtained as the optimal solution. The experimental results on simulated and real fMRI data sets show that the new method can not only accurately identify the connections and directions of the brain networks, but also quantitatively describe the connection strength of the brain networks, which has a good clinical application prospects.

Identifying functional modules in protein-protein interaction (PPI) networks is fundamental to understand cellular organization, processes, and functions. As an emerging evolutionary computational technology, swarm intelligence approaches are now becoming a new research hotspot in identifying functional modules. This paper proposes a new computational approach based on bacterial biological mechanisms for functional module detection in PPI networks (called as BBM-FMD). In BBM-FMD, each bacterium is first initialized to a candidate module partition by a random walk behavior. Then four biological mechanisms of bacteria including chemotaxis, conjugation, reproduction, and elimination and dispersal are simulated to iteratively search for better protein module partitions. At last, two post-processing steps are carried out to refine the obtained module partition. The experimental results on two PPI datasets demonstrate the superior performance of BBM-FMD in detecting functional modules compared with several other algorithms.

Multiobjective optimization problems widely exist in engineering application and science research. This paper presents an archive bacterial foraging optimizer to deal with multiobjective optimization problems. Under the concept of Pareto dominance, the proposed algorithm uses chemotaxis, conjugation, reproduction and elimination-and-dispersal mechanisms to approximate to the true Pareto fronts in multiobjective optimization problems. In the optimization process, the proposed algorithm incorporates an external archive to save the nondominated solutions previously found and utilizes the crowding distance to maintain the diversity of the obtained nondominated solutions. The proposed algorithm is compared with two state-of-the-art algorithms on four standard test problems. The experimental results indicate that our approach is a promising algorithm to deal with multiobjective optimization problems.

Community structure detection in complex networks contributes greatly to the understanding of complex mechanisms in many fields. In this article, we propose a multiagent evolutionary method for discovering communities in a complex network. The focus of the method lies in the evolutionary process of computational agents in a lattice environment, where each agent corresponds to a candidate solution to the community detection problem. First, the method uses a connection‐based encoding scheme to model an agent and a random‐walk behavior to construct a solution. Next, it applies three evolutionary operators, i.e., competition, crossover, and mutation, to realize information exchange among agents and solution evolution. We tested the performance of our method using synthetic and real‐world networks. The results show its capability in effectively detecting community structures.

Algorithms inspired by swarm intelligence have been used for many optimization problems and their effectiveness has been proven in many fields. We propose a new swarm intelligence algorithm for structural learning of Bayesian networks, BFO-B, based on bacterial foraging optimization. In the BFO-B algorithm, each bacterium corresponds to a candidate solution that represents a Bayesian network structure, and the algorithm operates under three principal mechanisms: chemotaxis, reproduction, and elimination and dispersal. The chemotaxis mechanism uses four operators to randomly and greedily optimize each solution in a bacterial population, then the reproduction mechanism simulates survival of the fittest to exploit superior solutions and speed convergence of the optimization. Finally, an elimination and dispersal mechanism controls the exploration processes and jumps out of a local optima with a certain probability. We tested the individual contributions of four algorithm operators and compared with two state of the art swarm intelligence based algorithms and seven other well-known algorithms on many benchmark networks. The experimental results verify that the proposed BFO-B algorithm is a viable alternative to learn the structures of Bayesian networks, and is also highly competitive compared to state of the art algorithms.

Detecting functional modules from a Protein-Protein Interaction (PPI) network is a fundamental and hot issue in proteomics research, where many computational approaches have played an important role in recent years. However, how to effectively and efficiently detect functional modules in large-scale PPI networks is still a challenging problem. We present a new framework, based on a multiple-grain model of PPI networks, to detect functional modules in PPI networks. First, we give a multiple-grain representation model of a PPI network, which has a smaller scale with super nodes. Next, we design the protein grain partitioning method, which employs a functional similarity or a structural similarity to merge some proteins layer by layer. Thirdly, a refining mechanism with border node tests is proposed to address the protein overlapping of different modules during the grain eliminating process. Finally, systematic experiments are conducted on five large-scale yeast and human networks. The results show that the framework not only significantly reduces the running time of functional module detection, but also effectively identifies overlapping modules while keeping some competitive performances, thus it is highly competent to detect functional modules in large-scale PPI networks.

Community structure detection in large-scale complex networks has been intensively investigated in recent years. In this paper, we propose a new framework which employs the ant colony clustering algorithm based on sampling to discover communities in large-scale complex networks. The algorithm firstly samples a small number of representative nodes from the large-scale network; secondly it uses the ant colony clustering algorithm to cluster the sampled nodes; thirdly it assigns the un-sampled nodes into the detected communities according to the similarity metric; finally it merges the initial clustering result to sustainably increase the modularity function value of the detection results. A significant advantage of our algorithm is that the sampling method greatly reduces the scale of the problem. Experimental results on computer-generated and real-world networks show the efficiency of our method.

BACKGROUND: Studies of functional modules in a Protein-Protein Interaction (PPI) network contribute greatly to the understanding of biological mechanisms. With the development of computing science, computational approaches have played an important role in detecting functional modules. RESULTS: We present a new approach using multi-agent evolution for detection of functional modules in PPI networks. The proposed approach consists of two stages: the solution construction for agents in a population and the evolutionary process of computational agents in a lattice environment, where each agent corresponds to a candidate solution to the detection problem of functional modules in a PPI network. First, the approach utilizes a connection-based encoding scheme to model an agent, and employs a random-walk behavior merged topological characteristics with functional information to construct a solution. Next, it applies several evolutionary operators, i.e., competition, crossover, and mutation, to realize information exchange among agents as well as solution evolution. Systematic experiments have been conducted on three benchmark testing sets of yeast networks. Experimental results show that the approach is more effective compared to several other existing algorithms. CONCLUSIONS: The algorithm has the characteristics of outstanding recall, F-measure, sensitivity and accuracy while keeping other competitive performances, so it can be applied to the biological study which requires high accuracy.

A protein-protein interaction (PPI) network is a biomolecule relationship network that plays an important role in biological activities. Studies of functional modules in a PPI network contribute greatly to the understanding of biological mechanism. With the development of life science and computing science, a great amount of PPI data has been acquired by various experimental and computational approaches, which presents a significant challenge of detecting functional modules in a PPI network. To address this challenge, many functional module detecting methods have been developed. In this survey, we first analyze the existing problems in detecting functional modules and discuss the countermeasures in the data preprocess and postprocess. Second, we introduce some special metrics for distance or graph developed in clustering process of proteins. Third, we give a classification system of functional module detecting methods and describe some existing detection methods in each category. Fourth, we list databases in common use and conduct performance comparisons of several typical algorithms by popular measurements. Finally, we present the prospects and references for researchers engaged in analyzing PPI networks.

Component-Based Software Engineering (CBSE) with Open Source Software and Commercial-Off-the-Shelf (COTS) components, Open Source Software (OSS) based development, and Software Outsourcing (SO) are becoming increasingly important for the Chinese software industry. It is therefore necessary to establish pragmatic and possibly nation-specific guidelines for Chinese software companies regarding the use of CBSE, OSS, and SO. Such guidelines should be based on insights from actual practice, which are in our case, obtained through surveys. A European state-of-the-practice survey on COTS- and OSS-oriented CBSE was conducted in Germany, Italy, and Norway in 2004-2005. We repeated similar surveys in China, with an extended survey on OSS and SO. We encountered many difficulties in conducting the surveys, but in most cases managed to find working solutions. We report on the lessons learned while conducting these surveys. In particular, we address issues relating to sampling, contacting respondents, data collection, and data validation. The main lessons are: 1) it was necessary to cooperate with a third-party organization with close relations to Chinese software companies; 2) it was necessary to assign researchers to this third-party organization to facilitate data collection and to control the quality of the data collected; and 3) an email survey, after an initial telephone call to establish contact, was the best method for getting questionnaires completed by Chinese respondents.

The business relationship between a software outsourcer and its suppliers is gradually moving from contract relationship to partnership. The partnership type between the outsourcer and the supplier is considered as a key predictor of outsourcing success. Although several studies have investigated the practices and benefits of building partnership from an outsourcer's perspective, few of them have studied these issues from the supplier's viewpoint, especially in the context of offshore software outsourcing. Since more and more Chinese software companies are getting outsourcing subcontracts from abroad, it is important to investigate the effect of business relationship on their performance, and to identify possible enhancements. Our study has collected data by a questionnaire-based survey from 53 finished projects in 41 Chinese software suppliers. Twenty-six of our investigated suppliers claim to have contract relationship with their outsourcers and the remaining 27 think they have partnership with outsourcers. Results from our study show, however, that 1) processes and methods used by suppliers to solve conflicts with outsourcers do not follow their self-claimed contract relationships or partnership; 2) there is no significant correlations between the type of relationship and the success of outsourced projects; 3) more personnel with proper language and communication skills need to be educated in order to facilitate Chinese companies to build and maintain a proper partnership with their outsourcers.

Ant colony optimization (ACO) algorithm has been applied to data mining recently. Aiming at Ant Miner, a classification rule learning algorithm based on ACO, this paper presents an enhanced Ant Miner, which includes two main contributions. Firstly, a rule punishing operator is employed to reduce the number of rules and the number of conditions. Secondly, an adaptive state transition rule and a mutation operator are applied to the algorithm to speed up the convergence rate. The results of experiments on some data sets demonstrate that the enhanced Ant-Miner can quickly discover better classification rules which have roughly competitive predicative accuracy and short rules

Personalized recommendation needs powerful Web Intelligence (WI) technologies to manage, analyze and employ various business data on the Web for e-business intelligence. This paper presents a novel recommendation framework on the Web, which is based on a multilevel customer model comprising three submodels, namely, the customer shopping model (CSM), the customer preference model (CPM), and the customer consumption model (CCM). These models capture a customer's information from different aspects. After preprocessing of raw data, we first build the CSM based on Bayesian networks by mining from customer shopping transactions, and then find the CPM by analyzing customer shopping history. Furthermore, the customer purchasing power can be formalized as a linear CCM. By combining the CSM with the present customer shopping action, a recommendation algorithm based on Bayesian probability inference is used to generate an individual recommendation set of commodities. A personalized filter including customization of the CPM and orientation of the CCM is also used to realize a more personalized recommendation. Experimental evaluation on real world data shows that the proposed approach can achieve personalized commodities recommendation efficiently and effectively.

Web Intelligence (WI) is a new and active research field in current AI and IT. Personalized recommendation in an intelligent B2C portal is an important research topic in WI. In this paper, we first investigate the architecture of a B2C portal from the viewpoint of conceptual levels of WI. Aiming at data mining of knowledge-level in a B2C portal, we present a new improved learning algorithm of Bayesian Networks, which consists of two major contributions, namely, making the best of lower order Conditional Independence (CI) tests and accelerating search process by means of sort order for parent nodes. By a number of experiments on ALARM datasets, we find that the proposed algorithm is both more efficient and effective than others. We have applied this algorithm to a commodity recommendation system in a B2C portal. Our experimental results demonstrate that the recommendation method based on a Customer Shopping Model (CSM) produced by the new algorithm outperforms some traditional ones in rates of coverage and precision.

As e-commerce developing rapidly, it is becoming a research focus about how to capture or find customer's behavior patterns and realize commerce intelligence by use of Web mining technology. Recommendation system in electronic commerce is one of the successful applications that are based on such mechanism. We present a new framework in recommendation system by finding customer model from business data. This framework formalizes the recommending process as knowledge representation of the customer shopping information and uncertainty knowledge inference process. In our approach, we firstly build a customer model based on Bayesian network by learning from customer shopping history data, then we present a recommendation algorithm based on probability inference in combination with the last shopping action of the customer, which can effectively and in real time generate a recommendation set of commodity.