Contents I. Introduction
In social commerce, good ratings and reviews of a product can increase the consumer’s trust which in turn influences consumer’s purchase intention [1]. To know the opinion of Thai consumers in the social commerce, Thai sentiment analysis or opinion mining can be used to automatically analyze the sentiments or opinions in Thai consumers’ reviews [2]. Trustworthiness of an e-vendor in Thailand can be measured in multiple dimensions such as product, price, and shipping dimensions. The consumer’s review can be automatically analyzed in multiple dimensions by using SenseComp method. In SenseComp, the sentiment in consumer’s review is analyzed by using multi-dimensional lexicon and sentiment compensation technique. The scores 1, -1, and 0 are given to the positive, negative, and neutral sentiments in the review respectively. The scores can be considered as the multi-dimensional trust score of e-vendor in e-marketplace and social commerce [3]. The weak point of SenseComp is the inability to give the correct scores to the reviews that have multiple dimensions and various sentiments in the same sentence. This is because there may be different sentiments describing the same dimension in the sentence.
