Sequential recommendation (SR) has attracted much research attention in the past few years. Most existing attribute integrated SR models do not directly model the complex relations between items and categorical attributes, as well do not exploit the power of attribute sequence in predicting the next item. In this paper, we propose an Item Categorical Attribute Integrated Sequential Recommendation (ICAI-SR) framework, which consists of an Item-Attribute Aggregation (IAA) model and Entity Sequential (ES) models. In IAA model, we employ a heterogeneous graph to represent the complex relations between items and different types of categorical attributes, then the attention mechanism based neighborhood aggregation is designed to model the correlations between items and attributes. For ES models, there are one Item Sequential (IS) model and one or more Attribute Sequential (AS) models. With IS and AS models, not only the item sequence but also the attribute sequence are used to predict the next item during model training. ICAI-SR is instantiated by taking Gated Recurrent Unit (GRU) and Bidirectional Encoder Representations from Transformers (BERT) as ES models, resulting in ICAI-GRU and ICAI-BERT respectively. Extensive experiments have been conducted on three public datasets to validate the performance of ICAI-SR. Experimental Results show that ICAI-SR performs better than both basic SR models and a competitive attribute integrated SR model.

Current state-of-the-art neural dialogue systems are mainly data-driven and are trained on human-generated responses. However, due to the subjectivity and open-ended nature of human conversations, the complexity of training dialogues varies greatly. The noise and uneven complexity of query-response pairs impede the learning efficiency and effects of the neural dialogue generation models. What is more, so far, there are no unified dialogue complexity measurements, and the dialogue complexity embodies multiple aspects of attributes—specificity, repetitiveness, relevance, etc. Inspired by human behaviors of learning to converse, where children learn from easy dialogues to complex ones and dynamically adjust their learning progress, in this paper, we first analyze five dialogue attributes to measure the dialogue complexity in multiple perspectives on three publicly available corpora. Then, we propose an adaptive multi-curricula learning framework to schedule a committee of the organized curricula. The framework is established upon the reinforcement learning paradigm, which automatically chooses different curricula at the evolving learning process according to the learning status of the neural dialogue generation model. Extensive experiments conducted on five state-of-the-art models demonstrate its learning efficiency and effectiveness with respect to 13 automatic evaluation metrics and human judgments.

Existing network embedding models regard all the links in a network as normal and model them without distinction. In real networks, there may be anomalous links like noise or adversarial links. We explicitly consider the existence of anomalous links in a network and propose anomaly aware network embedding (AANE) model. The key of AANE is the design of a new loss, which consists of anomaly aware loss and adjusted fitting loss. We adopt an anomaly indicator to iteratively select significant anomalous links from the network during model training, and removal loss and deviation loss are designed to model the reconstruction errors of selected anomalous and normal links respectively. To instantiate AANE, AAGAE and AAGCN are implemented on graph auto-encoder (GAE) and graph convolution based auto-encoder (GCNAE) respectively. For the purpose of evaluation, a heuristic anomalous link generation algorithm is proposed and by using the algorithm we generate anomalous links into six real world network datasets. Experimental results show that AANE outperforms both basic and competitive network embedding models in terms of anomalous link detection performance in most cases.

Topic modeling has become a widely used tool for document management. However, there are few topic models distinguishing the importance of documents on different topics. In this paper, we propose a framework LIMTopic to incorporate link based importance into topic modeling. To instantiate the framework, RankTopic and HITSTopic are proposed by incorporating topical pagerank and topical HITS into topic modeling respectively. Specifically, ranking methods are first used to compute the topical importance of documents. Then, a generalized relation is built between link importance and topic modeling. We empirically show that LIMTopic converges after a small number of iterations in most experimental settings. The necessity of incorporating link importance into topic modeling is justified based on KL-Divergences between topic distributions converted from topical link importance and those computed by basic topic models. To investigate the document network summarization performance of topic models, we propose a novel measure called log-likelihood of ranking-integrated document-word matrix. Extensive experimental results show that LIMTopic performs better than baseline models in generalization performance, document clustering and classification, topic interpretability and document network summarization performance. Moreover, RankTopic has comparable performance with relational topic model (RTM) and HITSTopic performs much better than baseline models in document clustering and classification.

The community and the topic can help summarize the text-augmented social networks. Existing works mixed up the community and the topic by regarding them as the same. However, there is an inherent difference between the community and the topic such that considering them as the same is not so flexible. We propose a mutual enhanced infinite (MEI) community–topic model to detect communities and topics simultaneously in text-augmented social networks. The community and the topic are correlated via a community–topic distribution. The mutual enhancement effect between the community and the topic is validated by introducing two novel measures perplexity with community (perplexityc) and mean of the rank (MRK) with topic (MRKt). To determine the numbers of communities and topics automatically, the Dirichlet Process Mixture (DPM) model and the Hierarchical Dirichlet Process mixture (HDP) model are used to model the community and the topic, respectively. We further introduce parameters to model the weight of the community and the topic responsible for community inference. Experiments on the co-author network built from a subset of DBLP data show that MEI outperforms the baseline models in terms of the generalization performance. Parameter study shows that MEI is averagely improved by 3.7 and 15.5% in perplexityc and MRKt, respectively, by setting a low weight for the topic. We also experimentally validate that MEI can determine the appropriate numbers of communities and topics.

Topic modeling has become a widely used tool for document management due to its superior performance. However, there are few topic models distinguishing the importance of documents on different topics. In this paper, we investigate how to utilize the importance of documents to improve topic modeling and propose to incorporate link based ranking into topic modeling. Specifically, topical pagerank is used to compute the topic level ranking of documents, which indicates the importance of documents on different topics. By retreating the topical ranking of a document as the probability of the document involved in corresponding topic, a generalized relation is built between ranking and topic modeling. Based on the relation, a ranking based topic model Rank Topic is proposed. With Rank Topic, a mutual enhancement framework is established between ranking and topic modeling. Extensive experiments on paper citation data and Twitter data are conducted to compare the performance of Rank Topic with that of some state-of-the-art topic models. Experimental results show that Rank Topic performs much better than some baseline models and is comparable with the state-of-the-art link combined relational topic model (RTM) in generalization performance, document clustering and classification by setting a proper balancing parameter. It is also demonstrated in both quantitative and qualitative ways that topics detected by Rank Topic are more interpretable than those detected by some baseline models and still competitive with RTM.