Assay: Hybrid Approach for Sentiment Analysis

  • D. V. Nagarjuna DeviEmail author
  • Thatiparti Venkata Rajini Kanth
  • Kakollu Mounika
  • Nambhatla Sowjanya Swathi
Conference paper
Part of the Smart Innovation, Systems and Technologies book series (SIST, volume 106)


We live in an age of massive online business, e-governance, and e-learning. All these activities involve transactions between customers, businessman, service providers, and recipients. Usually, the recipients give some comments on the quality of products and services. In this study, we proposed an algorithm named ASSAY (which means Analysis), to find the polarity at the document level. In our algorithm, initially we classify the reviews of each domain using naive Bayes and Support Vector Machine (SVM) algorithms which are in machine learning approach and then find the polarity at document level using HARN’s algorithm which comes under lexicon-based approach. In this algorithm, we use TextBlob for Parts of Speech (POS) tagging, where NV-Dictionary, ordinary dictionary, and SentiWordNet are used for extracting the polarities of features. Here, we combine both machine learning and lexicon-based approaches to calculate the result at document level accurately. In this way, we get the result about 80–85% more accurately than HARN’s algorithm which is proposed in lexicon-based approach.


NV-Dictionary Ordinary dictionary TextBlob 


  1. 1.
    Ahmad, S.R., Bakar, A.A., Yaakub, M.R.: Metaheuristic algorithms for feature selection in sentiment analysis. In: Science and information Conference 0152 July 28–30, 2015, London, UK (2015)Google Scholar
  2. 2.
    Rajashekar, P., Akhil, G.: Sentiment analysis using HARN’s algorithm. In: 2016 IEEE Conference (2016)Google Scholar
  3. 3.
    Khan, A., Baharudin, B.: Sentiment classification using sentence-level semantic orientation of opinion terms from blogs. In: IEEE National Postgraduate Conference (2011)Google Scholar
  4. 4.
    Hassan, S., Rafi, M., Shaikh, M.S.: Comparing SVM and naïve Bayes classifiers for text categorization with Wikitology as knowledge enrichment. In: 2011 IEEE 14th International Conference on Multitopic Conference (INMIC) (2011)Google Scholar
  5. 5.
    Pang, B., Vaithyanathan, S., Lee, L.: Thumbs up?. Sentiment classification using machine learning techniques, Empirical Methods in Natural Language Processing (2002)CrossRefGoogle Scholar
  6. 6.
    Pang, B., Lee, L.: A Sentimental Education: Sentiment Analysis Using Subjectivity Summarization Based on Minimum Cuts. The Association for Computational Linguistics, pp. 271–278 (2004)Google Scholar
  7. 7.
    WAD, W.A.A.: Machine learning algorithms in web page classification. Int. J. Comput. Sci. Informat. Technol. (IJCSIT) 4(5) (2012)Google Scholar
  8. 8.
    SentiWordNet. Last accessed 01 Sept 2018
  9. 9.
  10. 10.
    Lunando, E.: Indonesian social media sentiment analysis with sarcasm detection. In: 2013 International Conference on Advanced Computer Science and Information Systems (ICACSIS) (2013)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  • D. V. Nagarjuna Devi
    • 1
    Email author
  • Thatiparti Venkata Rajini Kanth
    • 2
  • Kakollu Mounika
    • 1
  • Nambhatla Sowjanya Swathi
    • 1
  1. 1.Rajiv Gandhi University of Knowledge TechnologiesNuzvidIndia
  2. 2.Jawaharlal Nehru Technological UniversityHyderabadIndia

Personalised recommendations