Analysis of Statistical Question Classification for Fact-Based Questions

Abstract

Abstract. Question classification systems play an important role in question answering systems and can be used in a wide range of other domains. The goal of question classification is to accurately assign labels to questions based on expected answer type. Most approaches in the past have relied on matching questions against hand-crafted rules. However, rules require laborious effort to create and often suffer from being too specific. Statistical question classification methods overcome these issues by employing machine learning techniques. We empirically show that a statistical approach is robust and achieves good performance on three diverse data sets with little or no hand tuning. Furthermore, we examine the role different syntactic and semantic features have on performance. We find that semantic features tend to increase performance more than purely syntactic features. Finally, we analyze common causes of misclassification error and provide insight into ways they may be overcome.

Towards a Protein–Protein Interaction information extraction system: Recognizing named entities

“…For this purpose, we define the head noun of a term as the last word in the term, if it does not contain any preposition, or as the last word before the preposition, otherwise. This approach is suitable to deal with compound terms in our CV, such as: cytoplasmic complementation assay whose head noun is assay; and colocalization by fluorescent probes cloning, whose head noun is colocalization, and is similar to that previously used by [60,61,62]. Head nouns appearing in more than five different concepts of an entity type are associated to it, and used during searching as explained below.…”

An Integrated Knowledge-based and Machine Learning Approach for Chinese Question Classification

“…We adopt SVM as the machine learning approach for CQC, motivated by the fact that they have consistently outperformed other machine learning techniques in several tasks, including text classification [11,13] and question classification [5,8,9,15,17].…”

“…Extensive works on question classification have been reported in the literature [1,4,5,8,9,10,12,14,15,17]. Early works suggested various standards for question classifications.…”

“…The goal of question classification is to accurately classify a question into a question type and then map it to an expected answer type (question type determination) [9]. For example, Chinese question classification for "奧運的發源地在哪裡？ (Where is the originating place of the Olympics?)"…”

“…Approaches to question classification (QC) can be considered in two broad classes, namely, rule-based and statistical [1,4,5,8,9,10,12,14,15,17]. In a rule-based approach, a domain expert produces a number of regular expressions and keywords.…”