Part-of-Speech tagging enhancement to natural language processing for Thai wh-question classification with deep learning

Abstract

Question classification is a crucial task for answer selection. Question classification could help define the structure of question sentences generated by features extraction from a sentence, such as who, when, where, and how. In this paper, we proposed a methodology to improve question classification from texts by using feature selection and word embedding techniques. We conducted several experiments to evaluate the performance of the proposed methodology using two different datasets (TREC-6 dataset and Thai sentence dataset) with term frequency and combined term frequency-inverse document frequency including Unigram, Unigram+Bigram, and Unigram + Trigram as features. Machine learning models based on traditional and deep learning classifiers were used. The traditional classification models were Multinomial Naïve Bayes, Logistic Regression, and Support Vector Machine. The deep learning techniques were Bidirectional Long Short-Term Memory (BiLSTM), Convolutional Neural Networks (CNN), and Hybrid model, which combined CNN and BiLSTM model. The experiment results showed that our methodology based on Part-of-Speech (POS) tagging was the best to improve question classification accuracy. The classifying question categories achieved with average micro 𝐹 1 -score of 0.98 when applied SVM model on adding all POS tags in the TREC-6 dataset. The highest average micro 𝐹 1 -score achieved 0.8 when applied GloVe by using CNN model on adding focusing tags in the Thai sentences dataset.

Effective Natural Language Processing and Interpretable Machine Learning for Structuring CT Liver-Tumor Reports

“…In this step, the above feature keywords are viewed as the semantic feature attributes. Accordingly, each training report is transformed into a semantic feature vector which is defined as: (7) where n_FK indicates the nth gram feature-keyword set and…”

An algorithm of test generation from functional specification using Open IE model and clustering

“…OpenIE model is used to build the syntax tree from test scenario sentence [7]. Before OpenIE processing, the text data should be prepared by the following algorithms: tokenization [10], lemmatization [11], part-of-speech definition [12], building the dependency tree D [13]. Triplets are formed with using of OpenIE according to the expression (1), where s is a subject, R is a relation, o is an object: 𝑇 𝑠 𝑅 𝑜 1 In some cases, an object contains a set of several interconnected natural language words.…”

A Survey on IBM Watson and Its Services

“…Syntactic are based on grammatical rules. Semantic analysis used name entity recognition (NET), natural language generation [4]. Example-" It is third highest mountain in world."…”

“…trains a computer by feeding at large amounts of data. These images can help to study "Melanorma" a form of skin cancer[4]. iii.Accelerating drug recovery: It takes very long time to introduce new drug to market and most experimental drug do not make it to market.…”