Title :
Is artificial immune system suitable for opinion mining?
Author :
Samsudin, Norlela ; Puteh, Mazidah ; Hamdan, Abdul Razak ; Nazri, Mohd Zakree Ahmad
Author_Institution :
Fac. of Comput. & Math. Sci., Univ. Teknol. MARA Terengganu, Dungun, Malaysia
Abstract :
Opinion mining is used to automate the process of identifying opinion whether it is a positive or negative view. Majority of previous works on this field uses natural language programming techniques to identify the sentiment. This paper reports the use of artificial immune system (AIS) technique in identifying Malaysian online movie reviews. This opinion mining process uses three string similarity functions namely Cosine Similarity, Jaccard Coefficient and Sorensen Coefficient. In addition, AIS performance was compared with other traditional machine learning techniques, which are Support Vector Machine, Naïve Baiyes and k-Nearest Network. The result of the findings are analyzed and discussed in this paper.
Keywords :
Bayes methods; Web sites; artificial immune systems; data mining; natural language processing; pattern classification; support vector machines; AIS; Jaccard coefficient; Malaysian online movie reviews; Sorensen coefficient; artificial immune system; cosine similarity; k-nearest network; naïve Baiyes; natural language programming techniques; opinion mining; string similarity functions; support vector machine; Data mining; Immune system; Motion pictures; Sociology; Statistics; Support vector machines; Training; artificial immune system; malaysian; movie reviews; opinion mining; sentiment mining;
Conference_Titel :
Data Mining and Optimization (DMO), 2012 4th Conference on
Conference_Location :
Langkawi
Print_ISBN :
978-1-4673-2717-6
DOI :
10.1109/DMO.2012.6329811