Nattapong Tongtep

An Empirical Study on Multi-dimensional Sentiment Analysis from User Service Reviews

Online reviews on a service are important sources for service providers to improve their service delivery and service consumers to obtain information for decision making before their service acquisition. However, in the real situation, there are several points of view (dimensions) in service assessment using online reviews. This paper shows an empirical study to apply classification-based sentiment analysis on online reviews with multiple dimensions using natural language processing techniques. The aim of this study is to find the most influential part-of-speech on the sentimental analysis and the performance of the multi-dimensional classification methods. By the experiments on reviews of restaurants with five dimensions, i.e., taste, environment, service, price, and cleanness, we find out that adjective (JJ) has the most influential part-of-speech on the sentimental analysis and BRplus is the most efficient one with the classification accuracy of 85.89%.

Simultaneous character-cluster-based word segmentation and named entity recognition in Thai language

Named entity recognition in inherent-vowel alphabetic languages such as Burmese, Khmer, Lao, Tamil, Telugu, Bali, and Thai, is difficult since there are no explicit boundaries among words or sentences. This paper presents a novel method to exploit the concept of character clusters, a sequence of inseparable characters, to group characters into clusters, utilize statistics among characters and their clusters to extract Thai words and then recognize named entities, simultaneously. Integrated of two phases, the word-segmentation model and the namedentity-recognition model, context features are exploited to learn parameters for these two discriminative probabilistic models, i.e., CRFs, to rank a set of word and named entity candidates generated. The experimental result shows that our method significantly increases the performance of segmenting word and recognizing entities with the F-measure of 96.14% and 83.68%, respectively.

A Feature-Based Approach for Relation Extraction from Thai News Documents

Relation extraction among named entities is one of the most important tasks in information extraction. This paper presents a feature-based approach for extracting relations among named entities from Thai news documents. In this approach, shallow linguistic processing, including pattern-based named entity extraction, is performed to construct several sets of features. Four supervised learning schemes are applied alternatively to investigate the performance of relation extraction using different feature sets. Focusing on four different types of relations in crime-related news documents, the experimental result shows that the proposed method achieves up to an accuracy of 95% using a data set of 1736 entity pairs. Effect of each set of features on relation extraction is explored for further discussion.

The monitoring management system for tangible cultural heritage surveillance

Cultural heritage surveillance is a primary mission of provincial cultural center affiliated to the Ministry of Culture. Formerly, the center received damage notification via letters, hotlines or directly investigating the cultural heritage sites. Thus, it is not comprehensive, prompt nor in time. This paper proposes a collaborative management system for monitoring and reporting unusual cases to the relevant cultural center for investigation. Related evidences with the correct geolocation are informed by users and recorded to the system. The petitions then will be automatically reported to the responsible authorities for prompt examination. The overall operation for monitoring and statistics are also provided.

Content-Based Readability Assessment: A Study Using A Syllabic Alphabetic Language (Thai)

Text readability is typically defined in terms of “grade level”; the expected educational level of the reader at which the text is directed. Mechanisms for measuring readability in English documents are well established; however this is not in case in many other languages, such as syllabic alphabetic languages. In this paper seven different mechanisms for assessing the readability of syllabic alphabetic language texts are proposed and compared. The mechanism are grouped under three headings: (i) graph ranking, (ii) document ranking, and (iii) hybrid. The presented comparison was conducted using the Thai language with respect to the reading age associated with secondary school, high school, and undergraduate students in the context of scientific abstract.