Search Result Personalization in Twitter Using Neural Word Embeddings

  • Sameendra Samarawickrama
  • Shanika Karunasekera
  • Aaron Harwood
  • Ramamohanarao Kotagiri
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10440)


In recent years, Twitter has become one of the most popular microblogging avenues. Today it has more than 300 million monthly active users generating more than 500 million tweets everyday. Twitter users both post messages as well as search for messages. Current search results given by Twitter are chronologically ordered and often users have to manually scan through an overwhelming number of the tweets to find content of interest. This process can quickly become infeasible. Personalization techniques address this problem by learning the user interests and tailoring search results by matching them with the user’s interests. Recent research on neural word embedding models, which represents each word in the vocabulary as a vector of real values, has gained much attention. These models learn word embeddings in such a way that contextually similar words have similar vectors. In this paper we propose a novel approach, PWEBA, for personalizing Twitter search, which uses neural word embeddings to model user interests. Our experimental results show that PWEBA outperforms existing approaches for all the evaluation metrics we have considered in this paper.


Twitter Personalization Content-mining Word-embeddings 


  1. 1.
    Bouadjenek, M.R., Hacid, H., Bouzeghoub, M.: Social networks and information retrieval, how are they converging? a survey, a taxonomy and an analysis of social information retrieval approaches and platforms. Inf. Syst. 56, 1–18 (2016)CrossRefGoogle Scholar
  2. 2.
    Carman, M.J., Baillie, M., Crestani, F.: Tag data and personalized information retrieval. In: Proceedings of the 2008 ACM Workshop on Search in Social Media, pp. 27–34. ACM (2008)Google Scholar
  3. 3.
    Culotta, A.: Training a text classifier with a single word using twitter lists and domain adaptation. Soc. Netw. Anal. Min. 6(1), 1–15 (2016)CrossRefGoogle Scholar
  4. 4.
    Dou, Z., Song, R., Wen, J.-R., Yuan, X.: Evaluating the effectiveness of personalized web search. TKDE 21(8), 1178–1190 (2009)Google Scholar
  5. 5.
    Efron, M.: Information search and retrieval in microblogs. J. Am. Soc. Inf. Sci. Technol. 62(6), 996–1008 (2011)CrossRefGoogle Scholar
  6. 6.
    Java, A., Song, X., Finin, T., Tseng, B.: Why we twitter: understanding microblogging usage and communities. In: WebKDD/SNA-KDD, pp. 56–65. ACM (2007)Google Scholar
  7. 7.
    Kang, J.H., Lerman, K.: Using lists to measure homophily on Twitter. In: AAAI Workshop on Intelligent Techniques for Web Personalization and Recommendation, pp. 26–32 (2012)Google Scholar
  8. 8.
    Lau, C.H., Li, Y., Tjondronegoro, D.: Microblog retrieval using topical features and query expansion. In: TREC (2011)Google Scholar
  9. 9.
    Leung, K.-T., Lee, D.L., Lee, W.-C.: Personalized web search with location preferences. In: ICDE, pp. 701–712. IEEE (2010)Google Scholar
  10. 10.
    Mikolov, T., Sutskever, I., Chen, K., Corrado, G.S., Dean, J.: Distributed representations of words and phrases and their compositionality. In: NIPS, pp. 3111–3119 (2013)Google Scholar
  11. 11.
    O’Connor, B., Krieger, M., Ahn, D.: Tweetmotif: exploratory search and topic summarization for twitter. In: ICWSM, pp. 384–385 (2010)Google Scholar
  12. 12.
    Rakesh, V., Singh, D., Vinzamuri, B., Reddy, C.K.: Personalized recommendation of twitter lists using content and network information. In: ICWSM, pp. 416–425 (2014)Google Scholar
  13. 13.
    Rzeszotarski, J.M., Spiro, E.S., Matias, J.N., Monroy-Hernández, A., Morris, M.R.: Is anyone out there? unpacking Q&A hashtags on twitter. In: CHI, pp. 2755–2758. ACM (2014)Google Scholar
  14. 14.
    Sontag, D., Collins-Thompson, K., Bennett, P.N., White, R.W., Dumais, S., Billerbeck, B.: Probabilistic models for personalizing web search. In: WSDM, pp. 433–442. ACM (2012)Google Scholar
  15. 15.
    Tao, K., Abel, F., Hauff, C., Houben, G.-J.: Twinder: a search engine for twitter streams. In: Brambilla, M., Tokuda, T., Tolksdorf, R. (eds.) ICWE 2012. LNCS, vol. 7387, pp. 153–168. Springer, Heidelberg (2012). doi: 10.1007/978-3-642-31753-8_11 CrossRefGoogle Scholar
  16. 16.
    Teevan, J., Ramage, D., Morris, M.R.: # twittersearch: a comparison of microblog search and web search. In: WSDM, pp. 35–44. ACM (2011)Google Scholar
  17. 17.
    Ushiama, T., Tominaga, K.: A method for personalized ranking of items based on similarity between twitter users. In: ICUIMC, pp. 44:1–44:4. ACM (2014)Google Scholar
  18. 18.
    Vallet, D., Cantador, I., Jose, J.M.: Personalizing web search with folksonomy-based user and document profiles. In: Gurrin, C., He, Y., Kazai, G., Kruschwitz, U., Little, S., Roelleke, T., Rüger, S., Rijsbergen, K. (eds.) ECIR 2010. LNCS, vol. 5993, pp. 420–431. Springer, Heidelberg (2010). doi: 10.1007/978-3-642-12275-0_37 CrossRefGoogle Scholar
  19. 19.
    Vosecky, J., Leung, K.W.-T., Ng, W.: Collaborative personalized twitter search with topic-language models. In: SIGIR, pp. 53–62. ACM (2014)Google Scholar
  20. 20.
    Wang, H., He, X., Chang, M.-W., Song, Y., White, R.W., Chu, W.: Personalized ranking model adaptation for web search. In: SIGIR, pp. 323–332. ACM (2013)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Sameendra Samarawickrama
    • 1
  • Shanika Karunasekera
    • 1
  • Aaron Harwood
    • 1
  • Ramamohanarao Kotagiri
    • 1
  1. 1.Department of Computing and Information SystemsThe University of MelbourneMelbourneAustralia

Personalised recommendations