Information extraction from scientific paper using rhetorical classifier

Abstract

Time constraints often lead a reader of scientific paper to read only the title and abstract of the paper, but reading these parts is often ineffective. This study aims to extract information automatically in order to help the readers get structured information from a scientific paper. The information extraction is done by rhetorical classification of each sentence in a scientific paper. Rhetoric information is the intention to be conveyed to the reader by the author of the paper. This research used corpusbased approach to build rhetorical classifier. Since there was a lack of rethorical corpus, we constructed our own corpus, which is a collection of sentences that have been labeled with rhetorical information. Each sentence represented as a vector of content, location, citation, and meta-discourses features. This collection of feature vectors is used to build rhetorical classifiers by using machine learning techniques. Experiments were conducted to select the best learning techniques for rhetorical classifier. Training set consists of 7239 labeled sentences, and the testing set consists of 3638 labeled sentences. We used WEKA (Waikato Environment for Knowledge Analysis) and LibSVM libraries. Learning techniques being considered were Naive Bayes, C4.5, Logistic, Multi-Layer Perceptron, PART, Instance-based Learning, and Support Vector Machines (SVM). The best performers are the SVM and Logistic classifier with accuracy of 0.51. By applying one-against-all strategy, the SVM accuracy can be improved to 0.60.

Rhetorical Sentences Classification Based on Section Class and Title of Paper for Experimental Technical Papers

“…Tokenization is the process of parsing a sentence or paragraph into units of words, referred to as tokens [3,13]. Each token will be a feature in the classification process [2]. This feature has a positive effect on accuracy.…”

“…These features consider the meaning of a word, such as its semantic features, and relations between words and sentences. An easy way to determine the relationship between sentences is classifying them based on rhetorical categories [2,16]. Teufel, et al [17] introduced 15 rhetorical categories in scientific papers, as shown in Table 1.…”

“…In other studies [1,2,15], there is no explanation of how sections were identified and classified. Our method uses Machine Learning to classify sections.…”

“…The main characteristic of a rhetorical sentence is that it contains a word-phrase that belongs to the author. A summary can be generated automatically by collecting rhetorical sentences based on sentence categories [1,2]. This research concerns the automatic extraction of rhetorical sentences based on four categories.…”

Automatic Tailored Multi-Paper Summarization based on Rhetorical Document Profile and Summary Specification

“…A rhetorical category is assigned for each sentence of the 75 papers in the collection. The result is an annotated corpus of 10877 rhetorically labeled sentences [35]. This corpus is then randomly split into a training set and a test set.…”