Profiling and Jury Selection using Sentiment Analysis
Rijuta Wagh1, Janvi Shah2, Khyati Shah3, Sindhu Nair4
1Rijuta Wagh, Department of Computer Engineering, Dwarkadas J. Sanghvi College of Engineering, Mumbai (Maharashtra), India.
2Janvi Shah, Department of Computer Engineering, Dwarkadas J. Sanghvi College of Engineering, Mumbai (Maharashtra), India.
3Khyati Shah, Department of Computer Engineering, Dwarkadas J. Sanghvi College of Engineering, Mumbai (Maharashtra), India.
4Prof. Sindhu Nair, Department of Computer Engineering, Dwarkadas J. Sanghvi College of Engineering, Mumbai (Maharashtra), India.
Manuscript received on 08 December 2015 | Revised Manuscript received on 16 December 2015 | Manuscript Published on 30 December 2015 | PP: 13-15 | Volume-5 Issue-7, December 2015 | Retrieval Number: G2239125715/2015©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Jury Selection is the process of selecting 12 jury members from a pool of random people. These selected Jurors attend the trial proceedings and after the closing statements give a verdict on whether the defendant is guilty or not. For a defendant to be pronounced guilty or not guilty the jurors must unanimously vote on it. If there isn’t a unanimous vote, then there is a mistrial. A mistrial can mean the whole case being restarted or the case being retired, meaning the case will not be pursued further. Thus the selection of the correct jurors is paramount to a decision in our favor, whichever side we may represent. We aim to develop a model in which the opinion of Twitter users is analyzed to create demographics which the lawyer can use for jury selection. Upon extracting data from Twitter based on hash tags pertaining to a certain case, the data undergoes an extensive cleaning process. We first classify the people according to age, sex, and profession and then plot graphs that can be statistically compared. This helps lawyers to make informed decisions and select a jury favorable to his/her case.
Keywords: Maximum Entropy, Naïve Bayes, Neural networks Sentiment Analysis, SVMs
Scope of the Article: Neural Networks Sentiment