Emotion Recognition in Conversation (ERC) has gained considerable attention due to its importance in human-computer interaction. In ERC task, the combination of multimodal information and contextual information is necessary since it can help the model understand emotional changes in the context from multiple perspectives. As Graph Neural Networks (GNNs) have shown the superiority in relation modeling, many graph-based methods have been proposed to improve the performance of emotion recognition by utilizing the edges to mine the contextual relationship and multimodal relationship in a conversation. However, the existence of numerous redundant edges and excessively complex modality interaction in the graph hinders the model from capturing the truly effective dependency information for emotion recognition. In this paper, we propose a Hypergraph based Contextual Relationship Modeling Method (HyperCRM) to carry out the ERC task. HyperCRM models a conversation as a hypergraph instead of a graph, which defines two types of hyperedges, namely speaker-level hyperedge and sequence-level hyperedge, to represent the contextual relationship within the same speaker and the local sequence of the conversation, respectively. Multimodal information is leveraged here as the node feature representation by the feature concatenation. In addition, an improved hypergraph convolution method is designed to capture the long-range contextual information by three-stage information propagation in the hypergraph, including node-hyperedge, hyperedge-hyperedge and node-hyperedge. The extensive experiments on two public datasets shows the new State-Of-The-Art (SOTA) results, to further demonstrate that the proposed method can simply make use of the multimodal information and effectively model the complex contextual relationships in the conversation.

The goal of domain adaptation (DA) is to develop a robust decision model on the source domain effectively generalize to the target domain data. State-of-the-art domain adaptation methods typically focus on finding an optimal inter-domain invariant feature representation or helpful instances from the source domain. In this paper, we propose a kernel extreme learning machine with discriminative transfer features and instance selection (KELM-DTF-IS) for unsupervised domain adaptation tasks, which consists of two steps: discriminative transfer feature extraction and classification with instance selection. At the feature extraction stage, we extend cross domain mean approximation(CDMA) by incorporating a penalty term and develop discriminative cross domain mean approximation (d-CDMA) to optimize the category separability between instances. Subsequently, d-CDMA is integrated into kernel ELM-AutoEncoder(KELM-AE) for extracting inter-domain invariant features. During the classification process, our approach uses CDMA metrics to compute a weights to each source instances based on their impact in reducing distribution differences between domains. Instances with a greater effect receive higher weights and vice versa. These weights are then used to distinguish and select source domain instances before incorporating them into weight KELM for proposing an adaptive classifier. Finally, we apply our approach to conduct classification experiments on publicly available domain adaptation datasets, and the results demonstrate its superiority over KELM and numerous other domain adaptation approaches.

Domain adaptation indeed promotes the progress of intelligent fault diagnosis in industrial scenarios. The abundant labeled samples are not necessary. The identical distribution between the training and testing datasets is not any more the prerequisite for intelligent fault diagnosis working. However, two issues arise subsequently: Feature learning in domain adaptation framework tends to be biased to the source domain, and unreliable pseudolabeling seriously impacts on the conditional domain adaptation. In this article, a new domain adaptation approach with self-supervised learning and feature clustering (DASSL-FC) is proposed, trying to alleviate the issues by unbiased feature learning and pseudolabels updating strategy. Taking different transformation methods as pretext, the transformed data and its pretext train a neural network in an SSL way. As to pseudolabeling, clusters are taken as the auxiliary information to correct the network predicted labels in terms of the "strong cluster" rule. Then, the updated pseudolabels and their confidence are enforced to further estimate the conditional distribution discrepancy and its confidence weight. To verify the effectiveness of the proposed method, the experiments are implemented on intraplatform and interplatforms for simulating the practical scenarios.

In real-world social networks, global information (e.g., the number of nodes and the connections between them) is incomplete or expensive to acquire; therefore, local community detection becomes especially important. Local community detection is used to identify the local community to which the given starting node belongs according to local information. For a given node, most existing local community detection methods can only find single scale local communities but not those of variable sizes. However, local communities with different scales are often required. Therefore, it is necessary and meaningful to find local communities of the given starting node with different scales; we call this multiscale local community detection. In this paper, we propose a new local modularity inspired by the global modularity and prove the equivalence of the proposed local modularity with two other typical local modularities. Furthermore, to detect local communities with different scales, we present a method based on the proposed local modularity. We test this method on several synthetic and real datasets, and the experimental results indicate that the detected community is meaningful and its scale can be changed reasonably.

Summary With the development of network technology, people are facing more and more massive information. How to extract emotional information in massive information rapidly has received more and more attention from people. This paper introduces the principle and structure of the traditional emotional model. Different personality, emotional states, and external stimuli will have different effects on emotional semantic analysis. In addition, this paper has proposed emotional semantic analysis method based on wake‐sleep and SVM method. The model starts from the description and calculation of the dynamic characteristics of emotions and more fully predicts the process characteristics that describe the evolution of emotions. Search and category browsing allows users to quickly access these information points. In addition, this paper provides a deep learning fusion algorithm in emotional semantic analysis, introduces its reference implementation and related key technologies, and supports business intelligence to a certain extent, and it has a strong application prospect on the network data information.

To meet the increasing requirements of HCI researchers who are prototyping a variety of forms to create novel interfaces under a ubiquitous situation, we present WeaveMesh, a low-fidelity and low-cost rapid prototyping system that produces 3D objects in a mesh structure. Inspired by hand-weaving craft, WeaveMesh supports a highly customizable software platform, which is applicable for simulating and facilitating freeform surface constructions composed of woven lines arranged in a regular grid, which can serve as a guide for easy assembly. In addition, mobilizable connectors are suggested to support flexible assembly, which can be revised, recycled, and reused to facilitate short iterations. Furthermore, compared to common additive and subtractive techniques, WeaveMesh has a better balance between time and material saving. In this paper, we will introduce the system in detail and demonstrate the feasibility of the technique through various 3D models in the area of interactive media, products and architecture.

With the increased usage of Internet, network security attracts many researchers to propose various kinds of approaches. Data mining techniques are efficient to construct a reliable Intrusion Detection System. Classification is an essential task in data mining. In this paper, a new classification method is proposed to build an accurate and efficient classifier for intrusion detection. The new classification method utilizes the average distances of the new data to its closest neighbor points to classify it as normal or intrusion. Then, the distances of the data to the centroids of normal, misuse intrusion and anomaly intrusion is used to get the accurate class label of the data. In addition, this paper integrates Fuzzy GNP-based class association rule mining method to extract rules. Fuzzy GNP avoids the use of the domain knowledge and solves the continuous attributes efficiently. On the basis of the extracted rules, the multi-feature space is projected into a two-dimensional average matching degree space. The benchmark data KDD Cup 1999 and NSL-KDD are used to evaluate the performance of the proposed method.

As the Internet services spread all over the world, many kinds and a large number of security threats are increasing. Therefore, intrusion detection systems, which can effectively detect intrusion accesses, have attracted attention. This paper describes a novel fuzzy class-association-rule mining method based on genetic network programming (GNP) for detecting network intrusions. GNP is an evolutionary optimization technique, which uses directed graph structures instead of strings in genetic algorithm or trees in genetic programming, which leads to enhancing the representation ability with compact programs derived from the reusability of nodes in a graph structure. By combining fuzzy set theory with GNP, the proposed method can deal with the mixed database that contains both discrete and continuous attributes and also extract many important class-association rules that contribute to enhancing detection ability. Therefore, the proposed method can be flexibly applied to both misuse and anomaly detection in network-intrusion-detection problems. Experimental results with KDD99Cup and DARPA98 databases from MIT Lincoln Laboratory show that the proposed method provides competitively high detection rates compared with other machine-learning techniques and GNP with crisp data mining.

With the rapid growth of the Internet, to make sure of the computer security has been a crucial problem, therefore, many techniques for Intrusion detection have been proposed in order to detect network attacks efficiently. On the other hand, data mining algorithms based on Genetic Network Programming (GNP) have been proposed recently. GNP is a graph-based evolutionary algorithm and can extract many important class association rules by making use of the distinguished representation ability of the graph structures. In this paper, a probabilistic classification is proposed and combined with the class association rule mining of GNP, and applied to Network intrusion detection for the performance evaluation. The proposed method creates a joint probability density function of normal and intrusion accesses and use it to efficiently classify new access data into normal, known intrusion or unknown intrusion. It is clarified from the experimental results that the proposed method shows high classification accuracy compared to the method without probabilistic classification.