Community detection is of crucial importance in understanding structures of complex networks. In many real-world networks, communities naturally overlap since a node usually has multiple community memberships. One popular technique to cope with overlapping community detection is Matrix Factorization (MF). However, existing MF-based models have ignored the fact that besides neighbors, "local non-neighbors" (e.g., my friend's friend but not my direct friend) are helpful when discovering communities. In this paper, we propose a Locality-based Non-negative Matrix Factorization (LNMF) model to refine a preference-based model by incorporating locality into learning objective. We define a subgraph called "k-degree local network" to set a boundary between local non-neighbors and other non-neighbors. By discriminately treating these two class of non-neighbors, our model is able to capture the process of community formation. We propose a fast sampling strategy within the stochastic gradient descent based learning algorithm. We compare our LNMF model with several baseline methods on various real-world networks, including large ones with ground-truth communities. Results show that our model outperforms state-of-the-art approaches.

Sentiment expression in microblog posts often reflects user's specific individuality due to different language habit, personal character, opinion bias and so on. Existing sentiment classification algorithms largely ignore such latent personal distinctions among different microblog users. Meanwhile, sentiment data of microblogs are sparse for individual users, making it infeasible to learn effective personalized classifier. In this paper, we propose a novel, extensible personalized sentiment classification method based on a variant of latent factor model to capture personal sentiment variations by mapping users and posts into a low-dimensional factor space. We alleviate the sparsity of personal texts by decomposing the posts into words which are further represented by the weighted sentiment and topic units based on a set of syntactic units of words obtained from dependency parsing results. To strengthen the representation of users, we leverage users following relation to consolidate the individuality of a user fused from other users with similar interests. Results on real-world microblog datasets confirm that our method outperforms state-of-the-art baseline algorithms with large margins.

Most word embedding methods take a word as a basic unit and learn embeddings according to words' external contexts, ignoring the internal structures of words. However, in some languages such as Chinese, a word is usually composed of several  characters and contains rich internal information. The semantic meaning of a word is also related to the meanings of its composing characters. Hence, we take Chinese for example, and present a character-enhanced word embedding model (CWE). In order to address the issues of character ambiguity and non-compositional words, we propose multiple-prototype character embeddings and an effective word selection method. We evaluate the effectiveness of CWE on word relatedness computation and analogical reasoning. The results show that CWE outperforms other baseline methods which ignore internal character information.

As compared to simple actions, activities are much more complex, but semantically consistent with a human's real life. Techniques for action recognition from sensor generated data are mature. However, there has been relatively little work on bridging the gap between actions and activities. To this end, this paper presents a novel approach for complex activity recognition comprising of two components. The first component is temporal pattern mining, which provides a mid-level feature representation for activities, encodes temporal relatedness among actions, and captures the intrinsic properties of activities. The second component is adaptive Multi-Task Learning, which captures relatedness among activities and selects discriminant features. Extensive experiments on a real-world dataset demonstrate the effectiveness of our work.

For particular sparse optimization problems, among them is the popular tasks, its low computation cost of linear subproblem proximal gradient (PG) based approach ([Beck and Teboulle, evaluation on each iteration leads to superior 2009]; [Nesterov, 2013]). This kind of methods can achieve practical performance. However, the inferior iteration the optimal rate of convergence under certain problem settings, complexity incurs excess number of gradient hence it enjoys low iteration complexity. The periteration evaluations, which can counteract the efficiency cost mainly comes from gradient evaluation and a gained by solving low cost linear subproblem. In proximal map (PM) related to the type of the regularizer. On this paper, we therefore propose a novel algorithm the one hand, due to the optimal iteration complexity, the that requires optimal graduate evaluations as proximal number of gradient evaluations is optimal for PG method.

In this paper we study the exchange of indivisible objects where agents’ possible preferences over the objects are strict and share a common structure among all of them, which represents a certain level of asymmetry among objects. A typical example of such an exchange model is a re-scheduling of tasks over several processors, since all task owners are naturally assumed to prefer that their tasks are assigned to fast processors rather than slow ones. We focus on designing exchange rules (a.k.a.mechanisms) that simultaneously satisfy strategyproofness, individual rationality, and Pareto efficiency. We first provide a general impossibility result for agents’ preferences that are determined in an additive manner, and then show an existence of such an exchange rule for further restricted lexicographic preferences. We finally find that for the restricted case, a previously known equivalence between the single-valuedness of the strict core and the existence of such an exchange rule does not carry over.

Crowdsourcing machine translation shows advantages of lower expense in money to collect the translated data. Yet, when compared with translation by trained professionals, results collected from non-professional translators might yield low-quality outputs. A general solution for crowdsourcing practitioners is to employ a large amount of labor force to gather enough redundant data and then solicit from it. Actually we can further save money by avoid collecting bad translations. We propose to score Turkers by their authorities during observation, and then stop hiring the unqualified Turkers. In this way, we bring both opportunities and risks in crowdsourced translation: we can make it cheaper than cheaper while we might suffer from quality loss. In this paper, we propose a graph-based PageRank-HITS Hybrid model to distinguish authoritative workers from unreliable ones. The algorithm captures the intuition that good translation and good workers are mutually reinforced iteratively in the proposed frame. We demonstrate the algorithm will keep the performance while reduce work force and hence cut cost. We run experiments on the NIST 2009 Urdu-to-English evaluation set with Mechanical Turk, and quantitatively evaluate the performance in terms of BLEU score, Pearson correlation and real money.

List of chairs of the 2015 IJCAI conference programs.

The triad is one of the most basic human groups in social networks. Understanding factors affecting the formation of triads will help reveal the underlying mechanisms that govern the emergence and evolution of complex social networks. In this paper, we study an interesting problem of decoding triadic closure in social networks. Specifically, for a given closed triad (a group of three people who are friends with each other), which link was created first, which followed, and which link closed. The problem is challenging, as we may not have any dynamic information. Moreover, the closure processes of different triads are correlated with each other. Our technical contribution lies in the proposal of a probabilistic factor graph model (DeTriad). The model is able to recover the dynamic information in the triadic closure process. It also naturally models the correlations among closed triads. We evaluate the proposed model on a large collaboration network, and the experimental results show that our method improves the accuracy of decoding triadic closure by up to 20% over that of several alternative methods.

Non-Intrusive Load Monitoring (NILM) uses one smart meter at the power feed to disaggregate the states of a set of appliances. Multiple NILM meters are deployed to achieve high monitoring accuracy in large-scale power systems. Our work studies the tradeoff between monitoring accuracy and meter deployment, in a quantitative and extensible way. In particular, we introduce a clearness function as an abstract indicator of expected monitoring accuracy given any NILM method, and then showcase two concrete constructions. With the notation of a clearness function, we propose solutions to the smart meter deployment problem (SMDP), that is, the problem of finding a deployment scheme with minimum number of meters while attaining a required monitoring accuracy. Theoretically, SMDP is shown NP-hard and a polynomial-time approximation scheme (PTAS) is proposed in this paper. For evaluation, we show that our proposed scheme is efficient and effective in terms of approximation ratio and running time. On real and simulated datasets, our proposed framework achieves a higher monitoring accuracy at a much lower cost, outperforming common baseline algorithms.