Jian-Tao Sun

Cross-domain sentiment classification via spectral feature alignment

Sentiment classification aims to automatically predict sentiment polarity (e.g., positive or negative) of users publishing sentiment data (e.g., reviews, blogs). Although traditional classification algorithms can be used to train sentiment classifiers from manually labeled text data, the labeling work can be time-consuming and expensive. Meanwhile, users often use some different words when they express sentiment in different domains. If we directly apply a classifier trained in one domain to other domains, the performance will be very low due to the differences between these domains. In this work, we develop a general solution to sentiment classification when we do not have any labels in a target domain but have some labeled data in a different domain, regarded as source domain. In this cross-domain sentiment classification setting, to bridge the gap between the domains, we propose a spectral feature alignment (SFA) algorithm to align domain-specific words from different domains into unified clusters, with the help of domain-independent words as a bridge. In this way, the clusters can be used to reduce the gap between domain-specific words of the two domains, which can be used to train sentiment classifiers in the target domain accurately. Compared to previous approaches, SFA can discover a robust representation for cross-domain data by fully exploiting the relationship between the domain-specific and domain-independent words via simultaneously co-clustering them in a common latent space. We perform extensive experiments on two real world datasets, and demonstrate that SFA significantly outperforms previous approaches to cross-domain sentiment classification.

Understanding user's query intent with wikipedia

Understanding the intent behind a users query can help search engine to automatically route the query to some corresponding vertical search engines to obtain particularly relevant contents, thus, greatly improving user satisfaction. There are three major challenges to the query intent classification problem: (1) Intent representation; (2) Domain coverage and (3) Semantic interpretation. Current approaches to predict the users intent mainly utilize machine learning techniques. However, it is difficult and often requires many human efforts to meet all these challenges by the statistical machine learning approaches. In this paper, we propose a general methodology to the problem of query intent classification. With very little human effort, our method can discover large quantities of intent concepts by leveraging Wikipedia, one of the best human knowledge base. The Wikipedia concepts are used as the intent representation space, thus, each intent domain is represented as a set of Wikipedia articles and categories. The intent of any input query is identified through mapping the query into the Wikipedia representation space. Compared with previous approaches, our proposed method can achieve much better coverage to classify queries in an intent domain even through the number of seed intent examples is very small. Moreover, the method is very general and can be easily applied to various intent domains. We demonstrate the effectiveness of this method in three different applications, i.e., travel, job, and person name. In each of the three cases, only a couple of seed intent queries are provided. We perform the quantitative evaluations in comparison with two baseline methods, and the experimental results shows that our method significantly outperforms other methods in each intent domain.

Query enrichment for web-query classification

Web-search queries are typically short and ambiguous. To classify these queries into certain target categories is a difficult but important problem. In this article, we present a new technique called query enrichment, which takes a short query and maps it to intermediate objects. Based on the collected intermediate objects, the query is then mapped to target categories. To build the necessary mapping functions, we use an ensemble of search engines to produce an enrichment of the queries. Our technique was applied to the ACM Knowledge Discovery and Data Mining competition (ACM KDDCUP) in 2005, where we won the championship on all three evaluation metrics (precision, F1 measure, which combines precision and recall, and creativity, which is judged by the organizers) among a total of 33 teams worldwide. In this article, we show that, despite the difficulty of an abundance of ambiguous queries and lack of training data, our query-enrichment technique can solve the problem satisfactorily through a two-phase classification framework. We present a detailed description of our algorithm and experimental evaluation. Our best result for F1 and precision is 42.4% and 44.4%, respectively, which is 9.6% and 24.3% higher than those from the runner-ups, respectively.

Web-page summarization using clickthrough data

Most previous Web-page summarization methods treat a Web page as plain text. However, such methods fail to uncover the full knowledge associated with a Web page needed in building a high-quality summary, because many of these methods do not consider the hidden relationships in the Web. Uncovering the hidden knowledge is important in building good Web-page summarizers. In this paper, we extract the extra knowledge from the clickthrough data of a Web search engine to improve Web-page summarization. Wefirst analyze the feasibility in utilizing the clickthrough data to enhance Web-page summarization and then propose two adapted summarization methods that take advantage of the relationships discovered from the clickthrough data. For those pages that are not covered by the clickthrough data, we design a thematic lexicon approach to generate implicit knowledge for them. Our methods are evaluated on a dataset consisting of manually annotated pages as well as a large dataset that is crawled from the Open Directory Project website. The experimental results indicate that significant improvements can be achieved through our proposed summarizer as compared to the summarizers that do not use the clickthrough data.

Thread detection in dynamic text message streams

Text message stream is a newly emerging type of Web data which is produced in enormous quantities with the popularity of Instant Messaging and Internet Relay Chat. It is beneficial for detecting the threads contained in the text stream for various applications, including information retrieval, expert recognition and even crime prevention. Despite its importance, not much research has been conducted so far on this problem due to the characteristics of the data in which the messages are usually very short and incomplete. In this paper, we present a stringent definition of the thread detection task and our preliminary solution to it. We propose three variations of a single-pass clustering algorithm for exploiting the temporal information in the streams. An algorithm based on linguistic features is also put forward to exploit the discourse structure information. We conducted several experiments to compare our approaches with some existing algorithms on a real dataset. The results show that all three variations of the single-pass algorithm outperform the basic single-pass algorithm. Our proposed algorithm based on linguistic features improves the performance relatively by 69.5% and 9.7% when compared with the basic single-pass algorithm and the best variation algorithm in terms of F1 respectively.

Feature selection in a kernel space

We address the problem of feature selection in a kernel space to select the most discriminative and informative features for classification and data analysis. This is a difficult problem because the dimension of a kernel space may be infinite. In the past, little work has been done on feature selection in a kernel space. To solve this problem, we derive a basis set in the kernel space as a first step for feature selection. Using the basis set, we then extend the margin-based feature selection algorithms that are proven effective even when many features are dependent. The selected features form a subspace of the kernel space, in which different state-of-the-art classification algorithms can be applied for classification. We conduct extensive experiments over real and simulated data to compare our proposed method with four baseline algorithms. Both theoretical analysis and experimental results validate the effectiveness of our proposed method.

A comparison of implicit and explicit links for web page classification

It is well known that Web-page classification can be enhanced by using hyperlinks that provide linkages between Web pages. However, in the Web space, hyperlinks are usually sparse, noisy and thus in many situations can only provide limited help in classification. In this paper, we extend the concept of linkages from explicit hyperlinks to implicit links built between Web pages. By observing that people who search the Web with the same queries often click on different, but related documents together, we draw implicit links between Web pages that are clicked after the same queries. Those pages are implicitly linked. We provide an approach for automatically building the implicit links between Web pages using Web query logs, together with a thorough comparison between the uses of implicit and explicit links in Web page classification. Our experimental results on a large dataset confirm that the use of the implicit links is better than using explicit links in classification performance, with an increase of more than 10.5% in terms of the Macro-F1 measurement.

Supervised latent semantic indexing for document categorization

Latent semantic indexing (LSI) is a successful technology in information retrieval (IR) which attempts to explore the latent semantics implied by a query or a document through representing them in a dimension-reduced space. However, LSI is not optimal for document categorization tasks because it aims to find the most representative features for document representation rather than the most discriminative ones. In this paper, we propose supervised LSI (SLSI) which selects the most discriminative basis vectors using the training data iteratively. The extracted vectors are then used to project the documents into a reduced dimensional space for better classification. Experimental evaluations show that the SLSI approach leads to dramatic dimension reduction while achieving good classification results.

Cross lingual text classification by mining multilingual topics from wikipedia

This paper investigates how to effectively do cross lingual text classification by leveraging a large scale and multilingual knowledge base, Wikipedia. Based on the observation that each Wikipedia concept is described by documents of different languages, we adapt existing topic modeling algorithms for mining multilingual topics from this knowledge base. The extracted topics have multiple types of representations, with each type corresponding to one language. In this work, we regard such topics extracted from Wikipedia documents as universal-topics, since each topic corresponds with same semantic information of different languages. Thus new documents of different languages can be represented in a space using a group of universal-topics. We use these universal-topics to do cross lingual text classification. Given the training data labeled for one language, we can train a text classifier to classify the documents of another language by mapping all documents of both languages into the universal-topic space. This approach does not require any additional linguistic resources, like bilingual dictionaries, machine translation tools, or labeling data for the target language. The evaluation results indicate that our topic modeling approach is effective for building cross lingual text classifier.

Multi-domain active learning for text classification

Active learning has been proven to be effective in reducing labeling efforts for supervised learning. However, existing active learning work has mainly focused on training models for a single domain. In practical applications, it is common to simultaneously train classifiers for multiple domains. For example, some merchant web sites (like Amazon.com) may need a set of classifiers to predict the sentiment polarity of product reviews collected from various domains (e.g., electronics, books, shoes). Though different domains have their own unique features, they may share some common latent features. If we apply active learning on each domain separately, some data instances selected from different domains may contain duplicate knowledge due to the common features. Therefore, how to choose the data from multiple domains to label is crucial to further reducing the human labeling efforts in multi-domain learning. In this paper, we propose a novel multi-domain active learning framework to jointly select data instances from all domains with duplicate information considered. In our solution, a shared subspace is first learned to represent common latent features of different domains. By considering the common and the domain-specific features together, the model loss reduction induced by each data instance can be decomposed into a common part and a domain-specific part. In this way, the duplicate information across domains can be encoded into the common part of model loss reduction and taken into account when querying. We compare our method with the state-of-the-art active learning approaches on several text classification tasks: sentiment classification, newsgroup classification and email spam filtering. The experiment results show that our method reduces the human labeling efforts by 33.2%, 42.9% and 68.7% on the three tasks, respectively.