Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Lateral computing wikipedia , lookup
Quantum machine learning wikipedia , lookup
Artificial intelligence wikipedia , lookup
Data analysis wikipedia , lookup
Learning classifier system wikipedia , lookup
Data assimilation wikipedia , lookup
Theoretical computer science wikipedia , lookup
International Conference on Computing and Mathematical Sciences – ICCMS’2017 February 25-26, 2017 Invent, Innovate and Integrate for Socioeconomic Development A Survey on Sentiment Analysis and Opinion Mining Iqra Kanwal1, Iqra Amjad2, Muhammad Sajid2 , Muhammad Shakeel Asghar2 Department of CS & IT University of Lahore, Gujrat Campus, Gujrat, Pakistan 2 Department of CS & IT University of Lahore, Gujrat Campus, Gujrat, Pakistan 1 [email protected], 2 [email protected], 2 [email protected], 2 [email protected] 1 Abstract—Now a day’s people get and share information through reviews or opinions via internet. Customer reviews containing very valuable information about different products and topics. There are different sources of data on web like social websites, discussion forums, and blogs containing such opinions. These sources mostly contain unstructured data need to be analyzed. This issue draws our attention to study the field of Opinion Mining and Sentiment Analysis. Opinion mining is basically Natural Language Processing and Information Extraction task which mine useful information and classify user reviews. Sentiment analysis is to classify reviews and assign polarity as positive, negative or neutral. In this paper we discuss different methods and techniques to classify reviews by using Machine Learning Techniques named as Supervised Learning techniques and Unsupervised Learning Techniques. In Supervised Machine Learning technique training data set is used classify sentence or document into finite set of classes, so we already know input and expected output. In Supervised Machine Learning Naïve-Bayes and Support-Vector Machine (SVM) K-nearest neighbor (KNN) is used at Sentence Level Sentiment Classification. . Maximum Entropy Classifier (ME) finds probabilistic model to satisfy any constraint, it converts label features into vectors by using encoding. In Unsupervised Machine Learning clustering algorithms are used for classification purpose training data set not used. Algorithms include K-means clustering and Hierarchical clustering. Different tools for sentiment analysis and opinion mining are NTLK, GATE, OPEN NLP, LING PIPE, and OPINION FINDER. Product/Service purchasing, marketing research, Recommendation system, spam Opinion Detection etc. are the application areas of sentiment analysis and opinion mining. 38