NEED TO ANALYZE SENTIMENTS OF MULTIPLE SOCIAL FORUMS USERS TO GAIN MEANINGFUL & ACTIONABLE INSIGHT Nikhil Govil 1, Saurabh Anand 2 1&2 Department of CEA, IET, GLA University, Mathura, U.P., (India) ABSTRACT Sentiment analysis or opinion mining is widely used in NLP (Natural Language Processing), text analysis and computational linguistics to reviewing or to identify and extract subjective information in source materials. The main objective of sentiment analysis is to determine the judgment or evaluation of a consumer, customer or a writer with respect to some topic. The judgment or evaluation may be affected by his / her emotional state or the past experience related to same type of product(s). In the same context, this paper shows the present scenario and the requirement to analyzing sentiments of users to get meaningful insight, on which analyzers may take proper corrective actions or decisions. Keywords: Big Data, NLP, Opinion Mining. I. INTRODUCTION Sentiment analysis or opinion analysis refers to the use of natural language processing, text analysis and computational linguistics to identify and extract subjective information in source materials. Sentiment analysis is widely applied to reviews and social media for a variety of applications, which may ranges from marketing to customer service. In this widely globalized and digital era, where most of the shopping is done on the internet, feedbacks are also expressed on the internet. Customers like to read reviews before going for purchasing any goods in these days. From mobile phone to movies, cars to holiday packages, everything has a review about it. If a product or service has a negative review, it also has been observed that it place a huge impact on next possible buyers. The person who posted these such types of reviews over internet or precisely said on social media might have a huge number of followers or have a sound friend list. By this way, his /her negative review may spread widely within seconds. In general, sentiment analysis aims to determine the attitude of a viewer or a writer with respect to some topic of a product or service. The attitude may be his or her judgment or evaluation, affective state (that is to say, the emotional state of the author when writing), or the intended emotional communication (that is to say, the emotional effect the author wishes to have on the reader). 441 P a g e
Several sources on the Web, provides deep insight about people s opinions on the products and services of various companies, social networking sites like Facebook, Twitter, Google+, LinkedIn, etc., blogs and discussion forums sent out loud messages through users, who voice their opinion openly. The data gathered from these forums or platforms is usually scattered and huge in size. Fig.1: Every Review is Important. [8] Because of these reasons, there is a need to design a framework through which accurately developed solution, where data from all the above sources on the web, trading back multiple years can be collected and processed to derive concise results. Our proposed work is supposed to extract expressions through these sources from multiple users. These opinions may have different subjects of various domains. I. WHAT IS BIG DATA? Big data is a broad term for data sets so large or complex that traditional data processing applications are inadequate. Analysis of data sets can find new correlations, to "spot business trends, prevent diseases, combat crime and so on." Scientists, business executives, practitioners of media, and advertising and governments alike regularly meet difficulties with large data sets in areas including Internet search, finance and business informatics. Scientists encounter limitations in e-science work, including meteorology, genomics, connectomics, complex physics simulations, and biological and environmental research. [7] Fig.2: Typical Uses of Big Data [7] 442 P a g e
Data sets grow in size in part because they are increasingly being gathered by cheap and numerous informationsensing mobile devices, aerial (remote sensing), software logs, cameras, microphones, radio-frequency identification (RFID) readers, and wireless sensor networks. Work with big data is necessarily uncommon; most analysis is of "PC size" data, on a desktop PC or notebook that can handle the available data set. Relational database management systems and desktop statistics and visualization packages often have difficulty handling big data. The work instead requires "massively parallel software running on tens, hundreds, or even thousands of servers". What is considered "big data" varies depending on the capabilities of the users and their tools, and expanding capabilities make big data a moving target. Thus, what is considered "big" one year becomes ordinary later? "For some organizations, facing hundreds of gigabytes of data for the first time may trigger a need to reconsider data management options. For others, it may take tens or hundreds of terabytes before data size becomes a significant consideration. [7] II. BENEFITS OF SENTIMENT ANALYSIS Following are few advantages to apply sentiment analysis: 1. Product perception: Gain insight of customer s sentiments and monitor change in trend over time. 2. Flame Detection: Evaluate procedures and processes which have given rise to negative sentiments. 3. Identify feedback sources to define new marketing targets and enhance visibility of product. 4. Reputation Management: Helps to innovate brand to enhance customer experience by offering tailor made solution and gain a competitive edge in the market. 5. Provides invaluable inputs in designing next generation products and services. III. CURRENT SCENARIO Most of the sentiments analysis tools available are merely depends on either 5 star or 10 star rating scale. In which user has to select any number of stars (out of either 5 or 10) as per his / her satisfaction about the product. This may not give the proper parameters to judge the exact opinion about the product until or unless user is not writing or recording his / her comments. These comments must be converted as parameters on which one can take better decision. 443 P a g e
Fig.3: A product s incomplete feedback to be analyzed. IV. TOOLS AVAILABLE By the course of time several tools have been developed for sentiments analysis consisting distinct features. Some of the most popular tools are as follows: 1. Meltwater: Assess the tone of the commentary as a proxy for brand reputation and uncover new insights that help you understand your target audience. 2. Google Alerts: A simple and very useful way to monitor your search queries. I use it to track content marketing and get regular email updates on the latest relevant Google results. This is a good starting point for tracking influencers, trends and competitors. 3. People Browser: Find all the mentions of your brand, industry and competitors and analyze sentiment. This tool allows you to compare the volume of mentions before, during and after your marketing campaigns. 4. Google Analytics: A powerful tool for discovering which channels influenced your subscribers and buyers. Create custom reports, annotations to keep uninterrupted records of your marketing and web design actions, as well as advanced segments to breakdown visitor data and gain valuable insights on their online experiences. 5. Hootsuite: A great freemium tool that allows you to manage and measure your social networks. The premium subscription provides enhanced analytics at a very reasonable 5.99 USD per month. 6. Tweetstats: This is a fun, free tool that allows you to graph your Twitter stats. Simply enter your Twitter handle and let the magic happen. 7. Facebook Insights: If you have more than 30 Likes on your Facebook Page you can start measuring its performance with Insights. See total page Likes, number of fans, daily active users, new Likes/Unlikes, Like sources, demographics, page views and unique page views, tab views, external referrers, media consumption and more! 8. Pagelever: This is another tool for measuring Facebook activity. Pagelever gives you the ability to precisely measure each stage of how content is consumed and shared on the Facebook platform. 444 P a g e
9. Social Mention: The social media equivalent to Google Alerts, this is a useful tool that allows you to track mentions for identified keywords in video, blogs, microblogs, events, bookmarks, comments, news, Q&A, hash tags and even audio media. It also indicates if mentions are positive, negative, or neutral. 10. Marketing Grader: Hubspot s Marketing Grader is a tool for grading your entire marketing funnel. It uses over 35 metrics to calculate your grade by looking at if you are regularly blog posting, Tweeting, updating on Facebook, converting visitors into leads, and more. It s a full funnel way to help you measure your inbound marketing initiatives. V. CONCLUSION In this paper, different sentiments issues of multi forum users are analyzed and discussed, the importance to develop a framework to do sentiments analysis electronically rather than manually which is undoubtedly an uphill task. Fig.4: Analysis of various reviews for better mining [8] In this context, our proposed work includes implementation of a framework for the opinion mining over provided big data which can suggest a meaningful and actionable insight. We hope that the framework can give improved understanding of scenarios as well as can present better option to the decision makers, analyst & managers to take the appropriate decisions for the betterment for their organization. REFERENCES [1] Raksha Sharma, Astha Agarwal, Mohit Gupta and Pushpak Bhattacharyya, Adjective Intensity and Sentiment Analysis, EMNLP2015, Lisbon, Portugal, Sept 17-21, 2015. [2] Ankit Ramteke, Akshat Malu, Pushpak Bhattacharyya and Saketha Nath, Detecting Trunarounds in Sentiment Analysis: Thwarting, ACE 2013, Sofia, Bulgaria 4-9 August, 2013. [3] Bo Pang and Lillian Lee, A Sentimental Education: Sentiment Analysis Using Subjectivity 445 P a g e
Summarization Based on Minimum Cuts, IEEE, August, 2008. [4] Bo Pang, and Lillian Lee, Opinion Mining & Sentiments Analysis, Foundation of Trends in Information Retrieval, Vol.2, No. 1-2, 2008. [5] Subhabrata Mukherjee and Pushpak Bhattacharyya, Sentiment Analysis in Twitter with Lightweight Discourse Analysis, COLING-2012, Mumbai 10-14 Dec., 2012. [6] Seema Acharya, Subhasini Chellappan, Big Data and Analytics (Wiley, First Edition, 2015). [7] http://www.cogno-sys.com/big-data/big -data-sentiment-analysis/ [8] Sentiment Analysis on Big Data Machine Learning Approach, SPAN White Paper, 2014. 446 P a g e