Relevancy and Similarity Aware Drug Comment Classification Framework on Social Media Drug Posts

  • D. Krithika Renuka
  • B. Rosiline Jeetha
Conference paper


Social media offer superior platform for its clients to share their comments and opinions about their knowledge toward specific product. This attracts the medical investigators to acquire knowledge about drug products owing to user perspective. Moreover, gaining and analyzing knowledge about drug tweet status is extremely a complex talk. This is carried out in our previous investigation by introducing Dynamic Drug Data examination by Hybrid Transductive Support Vector Machine with Fuzzy C Means (DDDA-HTSVM-FCM) procedure whose primary objective is to carry out drug tweet classification to acquire users’ opinion. But the research lacks in terms of accuracy of opinion identification. Accuracy of classification might get diminished in prior work with the occurrence of more extraneous tweets aggregated in online. The classification recital might be lesser in the prevailing work with the arrival of huge volume of data with either labeled or unlabeled data. These crises are resolved in the anticipated research technique by introducing Relevancy and Similarity Aware Drug Comment Classification Framework (RSDCCF). In this research work, similarity-based filtering technique is provided to eliminate irrelevant reviews from a number of tweets accumulated online. It is performed by evaluating semantic similarity among verbal of sentences and score is allocated based on similarity level. Verbal with fewer score will be eradicated to raise learning performance and to diminish computation overhead which caused by processing irrelevant data. Then, Classification performance is enhanced with presence of mixed data by initiating ensemble classifier. Classifiers used in this work are Adaboost, SVM, Random Forest, and TSVM. The complete implementation of research technique is carried out in MATLAB simulation, and it is proven that anticipated technique offers optimal outcome associated with accurate prediction of user opinion in drug comments than prevailing work.


Drug posts Twitter Ensemble classification Irrelevant post removal Similarity calculation Verbal score 



Adverse drug events


Dynamic Drug Data examination


FDA Adverse Event Reporting System


Fuzzy C Means


Food and Drug Administration


Hybrid Transductive Support Vector Machine


Internet Relay Chat




Relevancy and Similarity Aware Drug Comment Classification Framework


Support vector machine


Transductive support vector machines


Uniform resource locators


  1. 1.
    Serrano-Guerrero J, Olivas JA, Romero FP, Herrera-Viedma E (2015) Sentiment analysis: a review and comparative analysis of web services. Inf Sci 311:18–38CrossRefGoogle Scholar
  2. 2.
    Ravi K, Ravi V (2015) A survey on opinion mining and sentiment analysis: tasks, approaches and applications. Knowl-Based Syst 89:14–46CrossRefGoogle Scholar
  3. 3.
    Khan FH, Bashir S, Qamar U (2014) TOM: twitter opinion mining framework using hybrid classification scheme. Decis Support Syst 57:245–257CrossRefGoogle Scholar
  4. 4.
    Mann E, Zepeda O, Soones T, Federman A, Leff B, Siu A, Boockvar K (2018) Adverse drug events and medication problems in “Hospital at Home” patients. Home Health Care Serv Q:1–10Google Scholar
  5. 5.
    Food and Drug Administration, HHS (2016) Deeming tobacco products to be subject to the Federal Food, drug, and cosmetic act, as amended by the family smoking prevention and tobacco control act; restrictions on the sale and distribution of tobacco products and required warning statements for tobacco products. Final rule. Fed Regist 81(90):28973–29106Google Scholar
  6. 6.
    Peddie D, Small SS, Badke K, Bailey C, Balka E, Hohl CM (2018) Adverse drug event reporting from clinical care: mixed-methods analysis for a minimum required dataset. JMIR Med Inform 6(2):e10248CrossRefGoogle Scholar
  7. 7.
    Heimbach I, Schiller B, Strufe T, Hinz O (2015) Content virality on online social networks: empirical evidence from Twitter, Facebook, and Google+ on German news websites. In: Proceedings of the 26th ACM conference on hypertext & social media. ACM, New York, pp 39–47CrossRefGoogle Scholar
  8. 8.
    Nikfarjam A, Gonzalez G (2011) Pattern mining for extraction of mentions of adverse drug reactions from user comments. In: Proceedings of the American Medical Informatics Association (AMIA) annual symposium, pp 1019–1026Google Scholar
  9. 9.
    Gurulingappa H, Fluck J, Hofmann-Apitius M, Toldo L (2011) Identification of adverse drug event assertive sentences in medical case reports. In: First international workshop on knowledge discovery and health care management (KD-HCM), European conference on machine learning and principles and practice of knowledge discovery in databases (ECML PKDD), pp 16–27Google Scholar
  10. 10.
    Yeleswarapu S, Rao A, Joseph T, Saipradeep VG, Srinivasan R (2014) A pipeline to extract drug-adverse event pairs from multiple data sources. BMC Med Inform Decis Mak 14(13):1–16Google Scholar
  11. 11.
    Freifeld CC, Brownstein JS, Menone CM, Bao W, Filice R, Kass-Hout T, Dasgupta N (2014) Digital drug safety surveillance: monitoring pharmaceutical products in twitter. Drug Saf 37(5):343–350CrossRefGoogle Scholar
  12. 12.
    Greene JA, Choudhry NK, Kilabuk E, Shrank WH (2011) Online social networking by patients with diabetes: a qualitative evaluation of communication with Facebook. J Gen Intern Med 26(3):287–292CrossRefGoogle Scholar
  13. 13.
    Denecke K, Nejdl W (2009) How valuable is medical social media data? Content analysis of the medical web. Inf Sci 179(12):1870–1880CrossRefGoogle Scholar
  14. 14.
    Lu Y, Zhang P, Liu J, Li J, Deng S (2013) Health-related hot topic detection in online communities using text clustering. PLoS One 8(2):e56221CrossRefGoogle Scholar
  15. 15.
    Weitzman ER, Cole E, Kaci L, Mandl KD (2011) Social but safe? Quality and safety of diabetes-related online social networks. J Am Med Inform Assoc 18(3):292–297CrossRefGoogle Scholar
  16. 16.
    Shrank WH, Choudhry NK, Swanton K, Jain S, Greene JA, Harlam B, Patel KP (2011) Variations in structure and content of online social networks for patients with diabetes. Arch Intern Med 171(17):1589–1591CrossRefGoogle Scholar
  17. 17.
    Zhang Y, He D, Sang Y (2013) Facebook as a platform for health information and communication: a case study of a diabetes group. J Med Syst 37(3):1–12CrossRefGoogle Scholar
  18. 18.
    Goeuriot L, Na JC, Min Kyaing WY, Khoo C, Chang YK, Theng YL, Kim JJ (2012) Sentiment lexicons for health-related opinion mining. In: Proceedings of the 2nd ACM SIGHIT international symposium on health informatics. ACM, New York, pp 219–226CrossRefGoogle Scholar
  19. 19.
    Bian J, Topaloglu U, Yu F (2012) Towards large-scale twitter mining for drug-related adverse events. In: Proceedings of the international workshop on Smart health and wellbeing. ACM, New York, pp 25–32Google Scholar
  20. 20.
    Chee BW, Berlin R, Schatz B (2011) Predicting adverse drug events from personal health messages. In AMIA annual symposium on American medical informatics association, 217–26Google Scholar
  21. 21.
    Yang CC, Jiang L, Yang H, Tang X (2012) Detecting signals of adverse drug reactions from health consumer contributed content in social media. In Proceedings of ACM SIGKDD workshop on health informatics, pp 1–8Google Scholar
  22. 22.
    Leaman R, Wojtulewicz L, Sullivan R, Skariah A, Yang J, Gonzalez G (2010) Towards internet-age pharmacovigilance: extracting adverse drug reactions from user posts to health-related social networks. In Proceedings of the workshop on biomedical natural language processing, pp 117–125Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • D. Krithika Renuka
    • 1
  • B. Rosiline Jeetha
    • 2
  1. 1.Department of Computer Science (PG)PSGR Krishnammal College for WomenCoimbatoreIndia
  2. 2.Department of Computer Science, Dr. N.G.P College of Arts and ScienceCoimbatoreIndia

Personalised recommendations