• DocumentCode
    3660792
  • Title

    The Senti Strength Calculator: Engineering the Sentiment from the Opinionated Text

  • Author

    Deepak Thakur;Tanya Gera;Jaiteg Singh

  • Author_Institution
    Sch. of Comput. Sci., Chitkara Univ., Rajpura, India
  • fYear
    2015
  • fDate
    4/1/2015 12:00:00 AM
  • Firstpage
    1103
  • Lastpage
    1108
  • Abstract
    Sentiment analysis is the study of classifying human´s sentiments, evaluations, attitudes, opinions about some topic, expressed in form of text or speech. It is said that information is transmitting every time in the universe. The mode for the same can be anything, whether in writing, speaking, performing actions, etc. When someone writes about something, there is something more than just a text, is associated with it. Many people express their opinion about products by writing reviews, helping others to take decision. We may find numerous of reviews on e-commerce sites. Reading every review manually is a time consuming process. Thus, a automated approach is needed which can classify the reviews into desired classes. The intended work is to capture that important information from the unstructured text. The goal is to derive the qualitative sentiment analysis from the reviews. The work proposed a system known as senti strength calculator. The senti strength calculator is capable of engineering the sentiment from the opinionated text. The proposed system aims at finding the sentiment score of the review and extracting the features which have been commented upon by the opinion holders. The FCL (features counting lattice) will be generated which will determine the popular aspects and efficient ranking among the aspects can be deduced. The relative comparison among the products can be performed based on the opinions.
  • Keywords
    "Feature extraction","Sentiment analysis","Lattices","Calculators","Semantics","Classification algorithms","Algorithm design and analysis"
  • Publisher
    ieee
  • Conference_Titel
    Communication Systems and Network Technologies (CSNT), 2015 Fifth International Conference on
  • Type

    conf

  • DOI
    10.1109/CSNT.2015.136
  • Filename
    7280091