Advertisement

A Hybrid Lexicon-Based Sentiment and Behaviour Prediction System

  • Sumit GuptaEmail author
  • Puja HalderEmail author
Conference paper
  • 139 Downloads
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 591)

Abstract

Text mining is the process of extracting and/or deriving high-quality information from unstructured text by properly structuring the raw text. The structured texts serve as ideal candidates for revealing the syntactic and semantic interpretations encapsulated in them. To do so, we need to employ different methods of text mining and text analytics. Text Analytics deals with the objective of evaluating and assessing text by the application of natural language processing and other linguistic-oriented analytical methods. Text-based sentiment analysis aims to determine the attitude and sentimental state of an author by analysing different tokens of the texts in terms of their polarity. The aim of this paper is to propose a hybrid Lexicon-based sentiment and behaviour prediction system which can help one to comprehend the sentimental as well as the behavioural context of the author. We have used two sets of lexicons, viz. SenticNet 4.0 Lexicon and our own manually created Profile Lexicon in order to assess the input text and to predict the sentiment conveyed by the text as well as to identify the behaviour of the author. Our system works fairly in case of predicting both sentiment and behaviour by offering an accuracy of approximately 90%. Such a system has immense potential in identifying the real intention of an author once the behavioural and sentimental patterns of an author are predicted consummately.

Keywords

Text mining Text analytics Behaviour mining SenticNet Profile Lexicon 

References

  1. 1.
    Peng F, Schuurmans D, Wang S, Keselj V (2003) Language independent authorship attribution using character level language model. In: Proceedings of the tenth conference on European chapter of the association for computational linguistics, vol 1, Budapest, Hungary. ACM, pp 267–274Google Scholar
  2. 2.
    Ghose A, Ipeirotis PG (2008) Estimating the socio-economic impact of product reviews: mining text and reviewer characteristics. NYU Stern research working paperGoogle Scholar
  3. 3.
    Ghose A, Ipeirotis PG (2010) Estimating the helpfulness and economic impact of product reviews: mining text and reviewer characteristics. IEEE Trans Knowl Data Eng 23(10):1498–1512. IEEECrossRefGoogle Scholar
  4. 4.
    Yadav MP, Feeroz M, Yadav VK (2012) Mining the customer behaviour using web usage mining in e-commerce. In: Third international conference on computing communication and networking technologies (ICCCNT), pp 1–5, Coimbatore, IndiaGoogle Scholar
  5. 5.
    He W (2013) Examining students online interaction in a live video streaming environment using data mining and text mining. J Comput Hum Behav 29(1):90–102. ElsevierCrossRefGoogle Scholar
  6. 6.
    Lu J (2016) Empirical approaches for human behaviour analytics. School of Engineering, Computing and Mathematical SciencesGoogle Scholar
  7. 7.
    Textile Apex webpage https://textileapex.blogspot.com/2014/08/human-Behaviour.html. Accessed 28 July 2018
  8. 8.
    Tiwary V (2012) Social processes and behavioural issues. Gullybaba Publishing House (P) LtdGoogle Scholar
  9. 9.
    Cambria E, Poria S, Bajpai R, Schuller B (2016) A semantic resource for sentiment analysis based on conceptual primitives. In: Proceedings of the 26th international conference on computational linguistics: technical papers (COLING 2016), pp 2666–2677, Osaka, JapanGoogle Scholar
  10. 10.
    Rafeeque PC (2014) Large scale short text analysis in Twitter to identify same wavelength communities. Faculty of Information and Communication Engineering, Anna UniversityGoogle Scholar
  11. 11.
    Rafeeque PC, Sendhilkumar S, Mahalaxmi GS (2014) Twitter sentiment analysis for large-scale data: an unsupervised approach. Cogn Comput 7(2):254–262. SpringerGoogle Scholar
  12. 12.
    Sorto M, Aasheim C, Wimmer H (2017) Feeling the stock market: a study in the prediction of financial markets based on news sentiment. In: Proceedings of the southern association for information systems conference (SAIS 2017), St. Simons Island, GA, USA, AISeLGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  1. 1.Department of Computer Science & EngineeringUniversity Institute of Technology, The University of BurdwanBurdwanIndia

Personalised recommendations