Authors: Si, Jiasheng | Guo, Linsen | Zhou, Deyu
Article Type: Research Article
Abstract: Storyline extraction aims to generate concise summaries of related events unfolding over time from a collection of temporally-ordered news articles. Some existing approaches to storyline extraction are typically built on probabilistic graphical models that jointly model the extraction of events and the storylines from news published in different periods. However, their parameter inference procedures are often complex and require a long time to converge, which hinders their use in practical applications. More recently, a neural network-based approach has been proposed to tackle such limitations. However, event representations of documents, which are important for the quality of the generated storylines, are …not learned. In this paper, we propose a novel unsupervised neural network-based approach to extract latent events and link patterns of storylines jointly from documents over time. Specifically, event representations are learned by a stacked autoencoder and clustered for event extraction, then a fusion component is incorporated to link the related events across consecutive periods for storyline extraction. The proposed model has been evaluated on three news corpora and the experimental results show that it outperforms state-of-the-art approaches with significant improvements. Show more
Keywords: Storyline extraction, event representation, neural network
DOI: 10.3233/IDA-195061
Citation: Intelligent Data Analysis, vol. 25, no. 3, pp. 589-603, 2021
Authors: Zhou, Deyu | Chen, Liangyu | Zhang, Xuan | He, Yulan
Article Type: Research Article
Abstract: Social media provides unprecedented opportunities for people to disseminate information and share their opinions and views online. Extracting events from social media platforms such as Twitter could help in understanding what is being discussed. However, event extraction from social text streams poses huge challenges due to the noisy nature of social media posts and dynamic evolution of language. We propose a generic unsupervised framework for exploring events on Twitter which consists of four major steps, filtering, pre-processing, extraction and categorization, and post-processing. Tweets published in a certain time period are aggregated and noisy tweets which do not contain newsworthy events …are filtered by the filtering step. The remaining tweets are pre-processed by temporal resolution, part-of-speech tagging and named entity recognition in order to identify the key elements of events. An unsupervised Bayesian model is proposed to automatically extract the structured representations of events in the form of quadruples < entity, keyword, date, location > and further categorize the extracted events into event types. Finally, the categorized events are assigned with the event type labels without human intervention. The proposed framework has been evaluated on over 60 million tweets which were collected for one month in December 2010. A precision of 78.01% is achieved for event extraction using our proposed Bayesian model, outperforming a competitive baseline by nearly 13.6%. Moreover, events are also clustered into coherence groups with the automatically assigned event type labels with an accuracy of 42.57%. Show more
Keywords: Social media, event extraction, bayesian model, unsupervised learning
DOI: 10.3233/IDA-160048
Citation: Intelligent Data Analysis, vol. 21, no. 4, pp. 849-866, 2017
Authors: Guo, Linsen | Zhou, Deyu | He, Yulan | Xu, Haiyang
Article Type: Research Article
Abstract: Storyline generation aims to produce a concise summary of related events unfolding over time from a collection of news articles. It can be cast into an evolutionary clustering problem by separating news articles into different epochs. Existing unsupervised approaches to storyline generation are typically based on probabilistic graphical models. They assume that the storyline distribution at the current epoch depends on the weighted combination of storyline distributions in the latest previous M epochs. The evolutionary parameters of such long-term dependency are typically set by a fixed exponential decay function to capture the intuition that events in more recent epochs have …stronger influence to the storyline generation in the current epoch. However, we argue that the amount of relevant historical contextual information should vary for different storylines. Therefore, in this paper, we propose a new Dynamic Dependency Storyline Extraction Model (D 2 SEM) in which the dependencies among events in different epochs but belonging to the same storyline are dynamically updated to track the time-varying distributions of storylines over time. The proposed model has been evaluated on three news corpora and the experimental results show that it outperforms the state-of-the-art approaches and is able to capture the dependency on historical contextual information dynamically. Show more
Keywords: Storyline extraction, dynamic dependency, topic model, event extraction
DOI: 10.3233/IDA-184448
Citation: Intelligent Data Analysis, vol. 24, no. 1, pp. 183-197, 2020
Authors: Zhang, Xu | Hu, Xiaoyu | Liu, Zejie | Xiang, Yanzheng | Zhou, Deyu
Article Type: Research Article
Abstract: Text-to-SQL, a computational linguistics task, seeks to facilitate the conversion of natural language queries into SQL queries. Recent methodologies have leveraged the concept of slot-filling in conjunction with predetermined SQL templates to effectively bridge the semantic gap between natural language questions and structured database queries, achieving commendable performance by harnessing the power of multi-task learning. However, employing identical features across diverse tasks is an ill-suited practice, fraught with inherent drawbacks. Firstly, based on our observation, there are clear boundaries in the natural language corresponding to SELECT and WHERE clauses. Secondly, the exclusive features integral to each subtask are inadequately emphasized …and underutilized, thereby hampering the acquisition of discriminative features for each specific subtask. In an endeavor to rectify these issues, the present work introduces an innovative approach: the hierarchical feature decoupling model for SQL query generation from natural language. This novel approach involves the deliberate separation of features pertaining to subtasks within both SELECT and WHERE clauses, further dissociating these features at the subtask level to foster better model performance. Empirical results derived from experiments conducted on the WikiSQL benchmark dataset reveal the superiority of the proposed approach over several state-of-the-art baseline methods in the context of text-to-SQL query generation. Show more
Keywords: Text-to-SQL, multi-task learning, discriminative features, feature decoupling
DOI: 10.3233/IDA-230390
Citation: Intelligent Data Analysis, vol. 28, no. 4, pp. 991-1005, 2024
Authors: Zhang, Xu | Xiang, Yanzheng | Liu, Zejie | Hu, Xiaoyu | Zhou, Deyu
Article Type: Research Article
Abstract: Code search, which locates code snippets in large code repositories based on natural language queries entered by developers, has become increasingly popular in the software development process. It has the potential to improve the efficiency of software developers. Recent studies have demonstrated the effectiveness of using deep learning techniques to represent queries and codes accurately for code search. In specific, pre-trained models of programming languages have recently achieved significant progress in code searching. However, we argue that aligning programming and natural languages are crucial as there are two different modalities. Existing pre-train models based approaches for code search do not …effectively consider implicit alignments of representations across modalities (inter-modal representation). Moreover, the existing methods do not take into account the consistency constraint of intra-modal representations, making the model ineffective. As a result, we propose a novel code search method that optimizes both intra-modal and inter-modal representation learning. The alignment of the representation between the two modalities is achieved by introducing contrastive learning. Furthermore, the consistency of intra-modal feature representation is constrained by KL-divergence. Our experimental results confirm the model’s effectiveness on seven different test datasets. This paper proposes a code search method that significantly improves existing methods. Our source code is publicly available on GitHub.1 Show more
Keywords: Code search, semantic alignment, semantic representations, contrastive learning, pre-trained models
DOI: 10.3233/IDA-230082
Citation: Intelligent Data Analysis, vol. 28, no. 3, pp. 807-823, 2024
