CMUQ-Hybrid: Sentiment Classification By Feature Engineering and Parameter Tuning

Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014), pages 181–185, Dublin, Ireland, August 23-24, 2014. CMUQ-Hybrid: Sentiment Classification By Feature Engineering and Parameter Tuning Kamla Al-Mannai1, Hanan Alshikhabobakr2, Sabih Bin Wasi2, Rukhsar Neyaz2, Houda Bouamor2, Behrang Mohit2 Texas A&M University in Qatar1, Carnegie Mellon University in Qatar 2 almannaika@hotmail.com1 {halshikh, sabih, rukhsar, hbouamor, behrang}@cmu.edu Abstract This paper describes the system we sub- mitted to the SemEval-2014 shared task on sentiment analysis in Twitter. Our sys- tem is a hybrid combination of two system developed for a course project at CMU- Qatar. We use an SVM classifier and cou- ple a set of features from one system with feature and parameter optimization frame- work from the second system. Most of the tuning and feature selection efforts were originally aimed at task-A of the shared task. We achieve an F-score of 84.4% for task-A and 62.71% for task-B and the sys- tems are ranked 3rd and 29th respectively. 1 Introduction With the proliferation of Web2.0, people increas- ingly express and share their opinion through so- cial media. For instance, microblogging websites such as Twitter1 are becoming a very popular com- munication tool. An analysis of this platform re- veals a large amount of community messages ex- pressing their opinions and sentiments on differ- ent topics and aspects of life. This makes Twit- ter a valuable source of subjective and opinionated text that could be used in several NLP research works on sentiment analysis. Many approaches for detecting subjectivity and determining polarity of opinions in Twitter have been proposed (Pang and Lee, 2008; Davidov et al., 2010; Pak and Paroubek, 2010; Tang et al., 2014). For instance, the Twitter sentiment analysis shared task (Nakov et al., 2013) is an interesting testbed to develop and evaluate sentiment analysis systems on social media text. Participants are asked to implement This work is licensed under a Creative Commons At- tribution 4.0 International Licence. Page numbers and pro- ceedings footer are added by the organisers. Licence details: http://creativecommons.org/licenses/by/4.0/ 1http://twitter.com a system capable of determining whether a given tweet expresses positive, negative or neutral sen- timent. In this paper, we describe the CMUQ- Hybrid system we developed to participate in the two subtasks of SemEval 2014 Task 9 (Rosenthal et al., 2014). Our system uses an SVM classifier with a rich set of features and a parameter opti- mization framework. 2 Data Preprocessing Working with tweets presents several challenges for NLP, different from those encountered when dealing with more traditional texts, such as newswire data. Tweet messages usually contain different kinds of orthographic and typographical errors such as the use of special and decorative characters, letter duplication used generally for emphasis, word duplication, creative spelling and punctuation, URLs, #hashtags as well as the use of slangs and special abbreviations. Hence, before building our classifier, we start with a preprocess- ing step on the data, in order to normalize it. All letters are converted to lower case and all words are reduced to their root form …

Content truncated to 500 words