Weiguang Qu

AMR Parsing with an Incremental Joint Model.

To alleviate the error propagation in the traditional pipelined models for Abstract Meaning Representation (AMR) parsing, we formulate AMR parsing as a joint task that performs the two subtasks: concept identification and relation identification simultaneously. To this end, we first develop a novel componentwise beam search algorithm for relation identification in an incremental fashion, and then incorporate the decoder into a unified framework based on multiple-beam search, which allows for the bi-directional information flow between the two subtasks in a single incremental model. Experiments on the public datasets demonstrate that our joint model significantly outperforms the previous pipelined counterparts, and also achieves better or comparable performance than other approaches to AMR parsing, without utilizing external semantic resources.

Automatic Metaphor Recognition Based on Semantic Relation Patterns

Focusing on Chinese subject-predicate constructions, this paper analyzes the limitations of Selectional-Preference based metaphor recognition and proposes a new metaphor recognition model which is based on Semantic Relation Patterns. The model constructs Semantic Relation Pattern by integrating six types of semantic relations between a subject head and other subject heads in a subject-predicate cluster which share the same predicate head, and then employs a SVM classifier for metaphor recognition. Experiments show that the model outperforms the Selectional-Preference based metaphor recognition model to a great extent, achieving an F-1 of 89% in metaphor recognition, about 37% higher than Selectional-Preference based model. Analysis shows that the model is able to account for lexicalized metaphors, truth-condition literality and other types of literality and metaphor failed in Selectional-Preference based models. More importantly, the model can be generalized to unknown predicate heads. Theoretically, the semantic-relation-pattern model can also be applied in all endocentric constructions such as verb-objects and adjective-nouns.

Semantic change computation: A successive approach

The prevalence of creativity in the emergent online media language calls for more effective computational approach to semantic change. Two divergent metaphysical understandings are found with the task: juxtaposition-view of change and succession-view of change. This paper argues that the succession-view better reflects the essence of semantic change and proposes a successive framework for automatic semantic change detection. The framework analyzes the semantic change at both the word level and the individual-sense level inside a word by transforming the task into change pattern detection over time series data. At the word level, the framework models the word’s semantic change with S-shaped model and successfully correlates change patterns with classical semantic change categories such as broadening, narrowing, new word coining, metaphorical change, and metonymic change. At the sense level, the framework measures the conventionality of individual senses and distinguishes categories of temporary word usage, basic sense, novel sense and disappearing sense, again with S-shaped model. Experiments at both levels yield increased precision rate as compared with the baseline, supporting the succession-view of semantic change.

A collocation-based WSD model: RFR-SUM

In this paper, the concept of Relative Frequency Ratio (RFR) is presented to evaluate the strength of collocation. Based on RFR, a WSD Model RFR-SUM is put forward to disambiguate polysemous Chinese word sense. It selects 9 frequently used polysemous words as examples, and achieves the average precision up to 92:50% in open test. It has compared the model with Naive Bayesian Model and Maximum Entropy Model. The results show that the precision by RFR-SUM Model is 5:95% and 4:48% higher than that of Naive Bayesian Model and Maximum Entropy Model respectively. It also tries to prune RFR lists. The results reveal that leaving only 5% important collocation information can keep almost the same precision. At the same time, the speed is 20 times higher.

Detecting Spam Community Using Retweeting Relationships --- A Study on Sina Microblog

Microblog marketing is a new trend in social media. Spammers have been increasingly targeting such platforms to disseminate spam and promoting messages. Unlike the past behaviors on traditional media, they connect and support each other to perform spam tasks on microblogs. Therefore existing methods cant be directly used for detecting spam community. In this paper, we examine the behaviors of spammers on Sina microblog, and obtain some observations about their activities rules. Then we extract content features from tweet text and behavior features from retweeting interactions, perform machine learning to build classification models and identify spammers on microblogs. We evaluate our generated feature set used for detecting spammers under three classification methods, including Naive Bayes, Decision Tree and SVM. Extensive experiments show that our proposed feature set can make the classifiers perform well, and the crawler program combining the SVM classifier can effectively detect spam community.

An enhanced short text categorization model with deep abundant representation

Short text categorization is a crucial issue to many applications, e.g., Information Retrieval, Question-Answering System, MRI Database Construction and so forth. Many researches focus on data sparsity and ambiguity issues in short text categorization. To tackle these issues, we propose a novel short text categorization strategy based on abundant representation, which utilizes Bi-directional Recurrent Neural Network(Bi-RNN) with Long Short-Term Memory(LSTM) and topic model to catch more contextual and semantic information. Bi-RNN enriches contextual information, and topic model discovers more latent semantic information for abundant text representation of short text. Experimental results demonstrate that the proposed model is comparable to state-of-the-art neural network models and method proposed is effective.

Emotional Classification of Chinese Idioms Based on Chinese Idiom Knowledge Base

Idioms are not only interesting but also distinctive in a language for its continuity and metaphorical meaning in its context. This paper introduces the construction of a Chinese idiom knowledge base by the Institute of Computational Linguistics at Peking University and describes an experiment that aims at the automatic emotion classification of Chinese idioms. In the process, we expect to know more about how the constituents in a fossilized composition like an idiom function so as to affect its emotional properties.

A graph-based approach for semantic similar word retrieval

Semantic relatedness or semantic similarity between words is an important basic issue for many Natural Language Processing (NLP) applications, such as sentence retrieval, word sense disambiguation, question answering, and so on. This research issue attracts many researchers, but most of studies focus on improving the effectiveness, i.e., applying kinds of techniques to improve precision (effectiveness) but not efficiency. To tackle the problem, we propose to address the efficiency issue, that how to efficiently find top-k most semantic similar words to the query for a given dataset. This issue is very important for real applications especially for current big data. Efficient graph-based approaches on searching top-k semantic similar words are proposed in this paper. The results demonstrate that the proposed model can perform significantly better than baseline method.

A Study on Metaphors in Idioms Based on Chinese Idiom Knowledge Base

In Chinese language, idioms are an essential part of its vocabulary and used in everyday expression. People like to use idioms for their power of expression, rhetoric skill and special effect, which are mainly created by the metaphors in most of the idioms. This paper introduces a tentative research on the idioms with metaphors based on the Chinese Idiom Knowledge Base(CIKB) by the Institute of Computational Linguistics at Peking University (ICL/PKU), in which the author expects to provide due help to research and applications on this topic. We believe that research as such will have benefit on NLP tasks like automatic metaphor recognition and processing, semantic role labeling etc. On the other hand, our work may also contribute to lexicography, Chinese linguistics study and teaching Chinese as a foreign language.

Construction and Application of the Knowledge Base of Chinese Multi-word Expressions

In a language, Multi-word Expressions (MWEs, also called “idiomatic expressions” or “set phrases”) are very common in everyday usage. Most linguists hold that MWEs be an inclusive concept that should consist of not only lexical units such as idioms, idiomatic expressions, xiehouyu, proper nouns, but also non-lexical units such as proverbs, maxims and adages. Even those that are statistically idiosyncratic are to be listed in MWEs. In NLP tasks like word segmentation and semantic role labeling remain a bottle-neck problem. Therefore, to construct a knowledge base for MWEs with relatively complete entries and tagged attributes will be an effective solution for the above-mentioned problem. This paper introduces relevant information about the construction and application of an MWE knowledge base by the Institute of Computational Linguistics at Peking University(ICL/PKU), in which the author expects to provide due help to research in this regard.