Finding semantic communities using network topology and contents together is a hot topic in community detection. Existing methods often use word attributes in an indiscriminate way to help finding communities. Through analysis we find that, words in networked contents often embody a hierarchical semantic structure. Some words reflect a background topic of the whole network with all communities, some imply the high-level general topic covering several topic-related communities, and some imply the high-resolution specialized topic to describe each community. Ignoring such semantic structures often leads to defects in depicting networked contents where deep semantics are not fully utilized. To solve this problem, we propose a new Bayesian probabilistic model. By distinguishing words from either a background topic or some two-level topics (i.e., general and specialized topics), this model not only better utilizes the networked contents to help finding communities, but also provides a clearer multiplex semantic community interpretation. We then give an efficient variational algorithm for model inference. The superiority of this new approach is demonstrated by comparing with ten state-of-the-art methods on nine real networks and an artificial benchmark. A case study is further provided to show its strong ability in deep semantic interpretation of communities.

Recent research on community detection focuses on learning representations of nodes using different network embedding methods, and then feeding them as normal features to clustering algorithms. However, we find that though one may have good results by direct clustering based on such network embedding features, there is ample room for improvement. More seriously, in many real networks, some statisticallysignificant nodes which play pivotal roles are often divided into incorrect communities using network embedding methods. This is because while some distance measures are used to capture the spatial relationship between nodes by embedding, the nodes after mapping to feature vectors are essentially not coupled any more, losing important structural information. To address this problem, we propose a general Markov Random Field (MRF) framework to incorporate coupling in network embedding which allows better detecting network communities. By smartly utilizing properties of MRF, the new framework not only preserves the advantages of network embedding (e.g. low complexity, high parallelizability and applicability for traditional machine learning), but also alleviates its core drawback of inadequate representations of dependencies via making up the missing coupling relationships. Experiments on real networks show that our new approach improves the accuracy of existing embedding methods (e.g. Node2Vec, DeepWalk and MNMF), and corrects most wrongly-divided statistically-significant nodes, which makes network embedding essentially suitable for real community detection applications. The new approach also outperforms other state-of-the-art conventional community detection methods.

In-memory data processing frameworks (e.g., Spark) make big data analysis greatly simpler and efficient. However, stragglers that take much longer to finish than other tasks significantly degrade performance. There exist multiple factors that cause stragglers, either from the hardware resource layer or application layer, e.g. hardware heterogeneity, interference, data locality and data skew. While state-of-the-art straggler mitigation techniques have presented partial solutions on data skew and data locality, we find that the other factors can also result in serious problems. We present Clio, a cross-layer interference-aware optimization system that can effectively mitigate stragglers for data processing frameworks. Clio supports the scheduling of both map and reduce tasks. It heuristically dispatches intermediate data in proportion to the actual computing ability of each worker node, which is estimated considering various straggler factors, to balance the completion times of tasks in a much finer way. We implement Clio in Apache Spark, and evaluate its performance using both synthetic and real datasets. Experiment results show that, Clio can speed up the execution of applications by up to 67%, compared with the existing algorithms.

Topic models have many important applications in fields such as Natural Language Processing. Topic embedding modelling aims at introducing word and topic embeddings into topic models to describe correlations between topics. Existing topic embedding methods use documents alone, which suffer from the topical fuzziness problem brought by the introduction of embeddings of semantic fuzzy words, e.g. polysemous words or some misleading academic terms. Links often exist between documents which form document networks. The use of links may alleviate this semantic fuzziness, but they are sparse and noisy which may meanwhile mislead topics. In this paper, we utilize community structure to solve these problems. It can not only alleviate the topical fuzziness of topic embeddings since communities are often believed to be topic related, but also can overcome the drawbacks brought by the sparsity and noise of networks (because community is a high-order network information). We give a new generative topic embedding model which incorporates documents (with topics) and network (with communities) together, and uses probability transition to describe the relationship between topics and communities to make it robust when topics and communities do not match. An efficient variational inference algorithm is then proposed to learn the model. We validate the superiority of our new approach on two tasks, document classifications and visualization of topic embeddings, respectively.

Service clustering is the foundation of service discovery, recommendation and composition. Most of the existing methods mainly use service attribute information and ignore the semantic-based invocation relationships among service users. In fact, mutual invocation relationships between services occur on operations of the corresponding services, while service attributes are the whole service description. Our main challenge may be to effectively combine these two kinds of data for service clustering. To address this issue, we propose a new probabilistic generative model which contains two closely connected parts, one characterizing operation community memberships by using operation invocation relationships, and the other characterizing service cluster memberships by utilizing service attributes. The correlations between these two parts are characterized by the relationships between operation communities and service clusters. To train this model, we provide a nested expectation-maximization algorithm. Experimental results show its superior performance over the existing methods for service clustering.

Using network topology and semantic contents to find topic-related communities is a new trend in the field of community detection. By analyzing texts in social networks, we find that topics in networked contents are often hierarchical. In most cases, they have a two-level semantic structure with general and specialized topics, to respectively denote common and specific interests of communities. However, the existing community detection methods ignore such a hierarchy and take all words used to describe node semantics from an identical perspective. This indiscriminate use of words leads to natural defects in depicting networked content in which the deep semantics is not fully utilized. To address this problem, we propose a novel probabilistic generative model. By distinguishing the general and specialized topics of words, our model not only can find community structures more accurately, but also provide two-level semantic interpretation for each community. We train the model by deriving an efficient inference method under the framework of variational expectation-maximization. We provide a case study to show the ability of our algorithm in deep semantic interpretability of communities. The superiority of our algorithm for community detection is further demonstrated in comparison with eight state-of-the-art algorithms on eight real-world networks.

Community detection is an important task in social network analysis. Existing methods typically use the topological information alone, and ignore the rich information available in the content data. Recently, some researchers have noticed that user profiles can also benefit to community detection, and hence the combination of topology and node contents has become a new hot topic. Some methods using both topology and content have been proposed. However, they often suffer from two drawbacks: 1) they cannot extract a potential deep representation of the network; 2) they cannot automatically weight different information sources with adequate balance parameters. To overcome these issues, we propose a deep integration representation (DIR) algorithm via deep joint reconstruction, which is motivated by the similarity between deep feedforward auto-encoders and spectral clustering in terms of matrix reconstruction. Thanks to spectral clustering which is one of the best community detection methods, the proposed new method is also good at community discovery task. In addition, DIR has further benefit because it not only provides a nonlinear and deep representation of the network, but also learns the most suitable balance between different components automatically. We compare the proposed new approach with nine state-of-the-art community detection methods on eight real relatively large networks. The experimental results show the definite superiority of this new approach.

Discovery of communities in networks is a fundamental data analysis task. Recently, researchers have tried to improve its performance by exploiting node contents, and further interpret the communities using the derived semantics. However, the existing methods typically assume that the communities are assortative (i.e. members of each group are mostly connected to other members of the same group), and are unable to find the generalized community structure, e.g. structures with either assortative or disassortative communities (i.e. vertices of the same group have most of their connections outside their group), or a combination. In addition, these methods often assume that the network topology and node contents share the same group memberships, and thus cannot perform well when the contents mismatch with network structure. Also, they are limited to using only one topic to interpret each community. To address these two issues, we propose a new generative probabilistic model which is learned by using a nested expectation-maximization algorithm. It describes the generalized communities (based on network) and the content clusters (based on contents) separately, and further explores and models their correlation to improve as much as possible each of the communities and clusters based on the other. By depicting and utilizing this correlation, our model is not only robust with respect to the above problems, but is also able to interpret each community using more than one topic, which provides richer explanations. We validate the robustness of this proposed new approach on an artificial benchmark, and test its interpretability using a case study analysis. We finally show its definite superiority for community detection by comparing with seven state-of-the-art algorithms on eight real networks.

Collaborative filtering has been extensively studied in the context of ratings prediction. However, industrial recommender systems often aim at predicting a few items of immediate interest to the user, typically products that (s)he is likely to buy in the near future. In a collaborative filtering setting, the prediction may be based on the user's purchase history rather than rating information, which may be unreliable or unavailable. In this paper, we present an experimental evaluation of various collaborative filtering algorithms on a real-world dataset of purchase history from customers in a store of a French home improvement and building supplies chain. These experiments are part of the development of a prototype recommender system for salespeople in the store. We show how different settings for training and applying the models, as well as the introduction of domain knowledge may dramatically influence both the absolute and the relative performances of the different algorithms. To the best of our knowledge, the influence of these parameters on the quality of the predictions of recommender systems has rarely been reported in the literature.

In practical applications, recognition accuracy is sometimes not the only criterion; capability to reject erroneous patterns might also be needed. We show that there is a trade-off between these two properties. An efficient solution to this trade-off is brought about by the use of different algorithms implemented in various modules, i.e. multi-modular architectures. We present a general mechanism for designing and training multi-modular architectures, integrating various neural networks into a unique pattern recognition system, which is globally trained. It is possible to realize, within the system, feature extraction and recognition in successive modules which are cooperatively trained. We discuss various rejection criteria for neural networks and multi-modular architectures. We then give two examples of such systems, study their rejection capabilities and show how to use them for segmentation. In handwritten optical character recognition, our system achieves performances at state-of-the-art level, but is eight times faster. In human face recognition, our system is intended to work in the real world.

Neural Networks are very efficient for real world applications. However, practical problems often arise which can hinder performances. We discuss here some of these problems: under-representation of classes, rejection of outliers and ambiguous patterns, and illustrate the issues raised through various applications.

A connectionist approach to automatic speaker identification based on the learning vector quantization (VQ) algorithm is presented. For each adherent to the identification system, a number of references is fixed. The algorithm is based on a nearest-neighbor principle, with adaptation through learning. The identification is realized by comparing to a given threshold the distance of the unknown utterance to the nearest reference. Preliminary tests run on a ten-speaker set show an identification rate of 97% for MFC coefficients. The identification system and database used and the results obtained for different combinations of parameters are given. The system is evaluated by comparing its performances with a Bayesian system.>

A threshold network that can memorize and retrieve a given set of patterns $X_s $ is presented. A theoretical framework is given to allow expression of this property in terms of the dynamics of the network; the patterns in $X_s $ must be attractive fixed points. We give sufficient conditions for this, using a concept of energy. Scaling laws for the storing capacity of random patterns are provided and tested in the case of random as well as nonrandom patterns, such as the letters of the alphabet.