Online learning for data streams has gained significant attention in recent years. Existing methods impose various limitations on the feature space, classes, and labels of data streams. In contrast, real-world data streams often exhibit dynamic feature spaces, missing or partial feature values, undefined classes, and inaccessible labels. These complexities limit the applicability of current online learning methods to real-world scenarios. To fill the gap, this study explores a new online learning problem and proposes a novel algorithm:OnlineLearning forFickle Data Streams withoutLabels (OLFL). Specifically, OLFL has a triple main idea: 1) It tackles dimensional turbulence through a completion strategy that collaborates with weight vectors, reducing the over-reliance on surviving features; 2) It implements known class classification and new class detection via the density of core points within the neighborhood of each instance; 3) It utilizes a constraint-based adaptive coefficient vector to update model. We evaluate OLFL on extensive real-world and synthetic datasets against strong baselines, covering fickle feature spaces, emerging classes, and varying missingness rates. We further conduct ablation studies and case analyses. The results consistently show that OLFL outperforms the baselines in both accuracy and efficiency.

Web services are widely utilized across domains such as cloud computing, mobile networks, and Web applications. Due to their single-function nature, these services are often composed into Mashups to achieve more comprehensive functionality. However, the rapid growth in the number and variety of Web services has made it increasingly difficult to identify suitable services for Mashup development. Web service recommendation systems have emerged as a solution to this service overload, supporting innovative practices within the service-oriented development paradigm. While existing methods emphasize recommendation accuracy and relevance, few approaches simultaneously consider the personalized requirements of the Mashup side and the complementary relationships on the service side, both of which are essential for reconstructing the Web service ecosystem’s value chain. To address this gap, we propose PCNet, a Personalized Complementary Network for service recommendation based on tensor decomposition. We conceptualize the interaction dynamics between Mashups and services, as well as coinvocation patterns among services, using a three-dimensional tensor. The RESCAL tensor decomposition technique is then applied to jointly learn these relationships and uncover personalized complementary relationships among services. In addition, we develop a complementary perception module that uses an attention mechanism to dynamically model a Mashup’s focus on different complementary relationships, extending them to higher orders. Experimental results on real-world Web service datasets demonstrate that PCNet significantly outperforms state-of-the-art baselines. The implementation of PCNet is publicly available at: https://github.com/MengMeng3399/PCNet.

Efficient discovery and composition of services from large-scale cloud API repositories are fundamental to Mashup development and the rapid evolution of service-oriented software. However, extreme interaction sparsity severely limits existing recommendation methods, hindering their ability to infer unobserved relationships and capture complex service dependencies. Most prior approaches rely on deterministic point estimation in latent spaces, which leads to several critical issues under sparse supervision: difficulty in modeling preference uncertainty, high susceptibility to overfitting, and static and passive incorporation of multi-source auxiliary information. To address these challenges, we propose MS$^{2}$-Diff, a conditional diffusion-based generative framework for service recommendation that reformulates recommendation as a conditional probabilistic denoising process. MS$^{2}$-Diff has a three-fold main idea: 1) it characterizes two levels and four types of priors that drive Mashup development, including structural homophily and collaborative reachability at the structural level, as well as functional alignment and intent alignment at the semantic level; 2) it introduces a guidance-driven active retrieval mechanism via inverted attention within the denoising process, which actively retrieves and reconstructs latent relational patterns from noisy representations under the control of multi-view priors; and 3) it adopts a step-aware adaptive fusion strategy that dynamically coordinates structural and semantic priors across different denoising stages, enabling complementary and collaborative effects throughout the generation process. Extensive experiments on two real-world benchmark datasets demonstrate that MS$^{2}$-Diff achieves competitive and overall superior performance across multiple evaluation metrics, while improving robustness in sparse and cold-start scenarios.

Cloud API recommender system has emerged as a promising solution to address the overload problem caused by the overwhelming growth of cloud APIs, aiming to improve software development efficiency. However, the incremental development nature of the service-oriented software necessitates an adaptive complementary cloud API recommender system, while existing systems purely focus on generating single-function, high-quality, and personalized recommendations based on retrieval content, quality of service, or user preferences, and overlook the practical scenarios that recommend high-order complementary cloud APIs for developers. This paper addresses this gap by providing a logical reasoning network for high-order complementary cloud API recommendation (LRN4HCAR). We first construct three relation graphs to characterize cloud API co-invocation, function co-occurrence, and substitute relations, capturing the strong complementarity, weak complementarity, and non-substitute relations among cloud APIs. Next, we develop a high-order complementary logical reasoning network with three sub-networks: strong complementary logical reasoning, weak complementary logical reasoning, and non-substitute logical reasoning. This network enables the recommendation of cloud APIs with high-order complementary relationships while excluding substitutes. Utilizing LRN4HCAR, we evaluate classic and SOTA baselines across various complementary recommendation scenarios on two real-world datasets. Experimental results demonstrate that LRN4HCAR outperforms others in all experimental settings for complementary cloud API recommendations. We also verify the ability of LRN4HCAR to handle substitute noise and improve the visibility of long-tail cloud APIs in the recommendation results. The implementation code has made publicly available athttps://github.com/hey-mem/LRN4HCAR.

With the proliferation of the digital economy, cloud API has become the best carrier for data transmission and capability replication in software development. In the service-oriented software development paradigm, developers select appropriate cloud APIs for mashup creation based on their practical needs, where complementarity is an important consideration. Complementarity is reflected in various implicit relationships within mashups and cloud APIs, making complementarity analysis of cloud API ecosystems a critical issue. However, existing studies have overlooked the complementarity, and the lack of application variety in the primary dataset ProgrammableWeb has added a threat to the validity and generalization. In this paper, we first build a dataset of two cloud API ecosystems: ProgrammableWeb centering on mashup creation and Huawei AppGallery centering on mobile app development. Then, we conduct statistical and graph analysis to compare the two ecosystems. Finally, we propose a modeling approach to quantify complementarity and perform an analysis to investigate the role of complementarity in cloud API ecosystems. The results reveal differences between mashups and mobile apps in their preferences for cloud APIs. It also highlights the potential of complementarity in addressing the long-tail problem in cloud API ecosystems and improving cloud API recommendations.

Online learning is critical for handling complex data streams in Big Data-related applications. This study explores a new online learning problem where both the features and labels are incomplete. Such incompleteness poses a critical challenge in determining the latent relationship between incomplete features and labels. Unfortunately, existing online learning methods only consider a few cases of incomplete feature spaces, such as trapezoidal, evolvable, and capricious data streams, limiting their applicability to this problem. To bridge this gap, this study proposes a novel algorithm ofOnlineLearning for Data Streams withIncompleteFeatures andLabels (OLIFL). OLIFL imposes no constraints on changing patterns of feature space and does not require all instances to be labeled with two-fold ideas. First, OLIFL explores the informativeness of individual features to update the classifier by dynamically maintaining global feature space and updating the informativeness matrix. Second, it estimates the label confidence of unlabeled instances to control their negative effects by limiting the error upper bound. Extensive experiments on benchmark datasets are conducted in five scenarios: three incomplete feature (trapezoidal, evolvable, and capricious) spaces, and two incomplete labels (only missing labels and missing both features and labels). In addition, we explore the sensitivity of the model to parameters, and its usability and response efficiency in handling concept drifts. The results show that OLIFL significantly outperforms its rivals. Moreover, we use OLIFL to classify a movie review task as real application verification.

With the growing popularity of API-driven multiservice application (mashup) development, the burgeoning web APIs have left developers drowning in the sea of web API selections. Matching developers with the most appropriate APIs is the key to improving user satisfaction and promoting more popular web applications. As a result, more and more researchers pay attention to web API recommender systems based on collaborative filtering. However, employing collaborative filtering to recommend APIs is challenging due to the severe sparsity of mashup and API interactions. To address this problem, we propose a probabilistic generative model, called the Binary-API Topic model (BAT), to parameterize mashups and APIs. Technically, BAT is equipped with a mechanism to extract binary-APIs and predict unknown pairwise interactions. To improve generality and capture more relevance from a limited number of interactions, we learn binary-API topics by directly modeling the generation of API co-occurrence patterns across the repository (all mashup collections from programmableweb.com). The main advantage of BAT is that it preserves API co-occurrence patterns in model learning and exploits the rich global relevance. Finally, through extensive experiments, we demonstrate that BAT can achieve the highest performance on the sparse real-world data set.

Temporal Action Detection(TAD) is a challenge task in video understanding. The current methods mainly use global features for boundary matching or predefine all possible proposals, while ignoring long context information and local action boundary features, resulting in the decline of detection accuracy. To fill this gap, we propose a Dilation Location Network (DL-Net) model to generate more precise action boundaries by enhancing boundary features of actions and aggregating long contextual information in this paper. Specifically, we design the boundary feature enhancement (BFE) block, which strengthens the actions boundary feature and fuses the similar feature of the different channels by pooling and channel squeezing. Meanwhile, in action location, we design multiple dilated convolutional structures to aggregate long contextual information of time point/interval. We conduct extensive experiments on ActivityNet-1.3 and Thumos14 show that DL-Net is capable of enhancing action boundary features and aggregating long contextual information effectively.

Utilizing traditional object detectors to build a few-shot object detection (FSOD) model ignores the differences between classification and regression tasks and causes task conflict and class confusion, resulting in a decline in classification performance. In contrast, this paper focuses on the above shortcomings and utilizes the strategies of Back Propagation and Dynamic Learning to construct a model for addressing FSOD, named BPDL. Our BPDL has a two-fold main idea: a) it uses the optimized localization boxes to alleviate the task conflict and refine classification features by a correction loss, and b) it develops a dynamic learning strategy to filter the confusing features and mine more realistic prototype representations of the categories to calibrate classification. Extensive experiments on multiple benchmarks show that our BPDL model outperforms existing methods and advances the FSOD task’s state-of-the-art.

Multi-label streaming feature selection has attracted extensive attention in diverse big data applications. However, most existing works focused on the scenarios where labels are independent, while ignoring the real scenarios that they may be interdependent and correlated with each other. This paper aims to fill this gap by developing a novel online multi-label streaming feature selection scheme by taking into account the existence of label correlation, known as (OMSFSLC). In our design, we first calculate the correlation degree between labels to obtain the label weight. Then, we integrate the mutual information and the label weight to evaluate the correlation between features and labels. In particular, it consists of three stages: 1) online significance analysis, which can determine the significant features via the correlation degree between the newly arriving features and labels; 2) online relevance analysis, which can obtain relevant features via the mutual information; and 3) online redundancy analysis, which can filter the redundant features for removal via pairwise comparison. We implement our solution and conduct extensive experiments on benchmark datasets for performance evaluations. The experimental results exhibit that OMSFSLCsignificantly outperforms the state-of-the-art methods in terms of effectiveness and efficiency.

Learning with streaming data has attracted extensive research interest in recent years. Existing online learning approaches have specific assumptions regarding data streams, such as requiring fixed or varying feature spaces with explicit patterns and balanced class distributions. While the data streams generated in many real scenarios commonly have arbitrarily incomplete feature spaces and dynamic imbalanced class distributions, making existing approaches be unsuitable for real applications. To address this issue, this paper proposes a novelOnlineLearning fromIncomplete andImbalancedDataStreams (OLI$^{2}$DS) algorithm. OLI$^{2}$DS has a two-fold main idea: 1) it follows the empirical risk minimization principle to identify the most informative features of incomplete feature spaces, and 2) it develops a dynamic cost strategy to handle imbalanced class distributions in real-time by transforming F-measure optimization into a weighted surrogate loss minimization. To evaluate OLI$^{2}$DS, we compare it with state-of-the-art related algorithms in three kinds of experiments. First, we adopt 14 real datasets to simulate three scenarios of incomplete feature spaces, i.e., trapezoidal, feature evolvable, and capricious data streams. Second, based on a benchmark online analyzer, we generate 13 datasets to simulate incomplete data streams with different imbalance ratios. Third, we analyze concept drift in two simulated scenes, i.e., online learning and data stream mining, and verify the adaption of OLI$^{2}$DS on repeated concept drifts and variable imbalance ratios. The results demonstrate that OLI$^{2}$DS achieves a significantly better performance than its rivals. Besides, a real-world case study on movie review classification is conducted to elaborate on our OLI$^{2}$DS algorithm's effectiveness. Code is released athttps://github.com/youdianlong/OLI2DS.

Recently, causal feature selection (CFS) has attracted considerable attention due to its outstanding interpretability and predictability performance. Such a method primarily includes the Markov blanket (MB) discovery and feature selection based on Granger causality. Representatively, the max-min MB (MMMB) can mine an optimal feature subset, i.e., MB; however, it is unsuitable for streaming features. Online streaming feature selection (OSFS) via online process streaming features can determine parents and children (PC), a subset of MB; however, it cannot mine the MB of the target attribute ( T ), i.e., a given feature, thus resulting in insufficient prediction accuracy. The Granger selection method (GSM) establishes a causal matrix of all features by performing excessively time; however, it cannot achieve a high prediction accuracy and only forecasts fixed multivariate time series data. To address these issues, we proposed an online CFS for streaming features (OCFSSFs) that mine MB containing PC and spouse and adopt the interleaving PC and spouse learning method. Furthermore, it distinguishes between PC and spouse in real time and can identify children with parents online when identifying spouses. We experimentally evaluated the proposed algorithm on synthetic datasets using precision, recall, and distance. In addition, the algorithm was tested on real-world and time series datasets using classification precision, the number of selected features, and running time. The results validated the effectiveness of the proposed algorithm.

Skeleton-based action recognition has achieved promising performance recently, but there are still many challenges, e.g., the structural relation between joints and the different attention of frames, due to the complex spatial-temporal evolution of skeletal joints. In this paper, we propose a spatial reasoning and context-aware attention network for skeleton-based action recognition, which consists of a spatial reasoning module and a context-aware attention module. The spatial reasoning module can exploit the structural relation between joints to obtain the spatial features within each skeleton frame, followed by the context-aware attention module learning the different attention of frames. We perform experiments on two datasets and verify the effectiveness of each module in the proposed network. The comparison results demonstrate that our method achieves state-of-the-art performance.

With the advent of service-oriented computing, Web service QoS prediction have become crucial for the success of many service-based distributed systems due to its significant role in evaluating Web services to satisfy user personalized needs. Previous studies underestimate the role of implicit relationship of neighborhood with respect to both users and Web services. It is natural and reasonable to assume that the QoS Web services perform and the choices users make are significantly influenced by their neighbors, thus it will be beneficial to incorporate this new source of information in QoS prediction. In this paper, we propose a neighborhood regularized matrix factorization method for collaborative Web service QoS prediction by properly incorporating both user and Web service neighborhood relationships aiming to improve the accuracy of prediction as well as alleviate the data sparsity issue. Experimental results demonstrate that the proposed approach achieves more accurate prediction than other well-known methods.