Abstract
While social media presence is increasingly important for businesses, growing a social media account and improving its reputation by gathering followers are time-consuming tasks, especially for professionals and small businesses lacking the necessary skills and resources. With the broader goal of providing automatic tool support for social media account automation, in this paper we consider the problem of recommending a Twitter account manager a top-K list of Twitter users that, if approached—e.g., followed, mentioned, or otherwise targeted on social media—are likely to follow the account and interact with it, this way improving its reputation. We propose a recommendation system tackling this problem that leverages features ranging from basic social media attributes to specialized, domain-relevant user profile attributes predicted from data using machine learning techniques, and we report on a preliminary analysis of its performance in gathering new followers in a Twitter scenario where the account manager follows recommended users to trigger their follow-back.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The decision has to occur in a reasonable time window after the user has been engaged (causality). For simplicity, we assume also that the account manager trusts the system and engages all recommended users. If not, then also the feedback by the account manager about whether or not to engage a user should be accounted for.
- 2.
In our tests, we spent 3 min per user if visual clues (e.g., profile picture and banner) could be exploited to assess a user’s relevancy, and up to 15 min if the full profile had to be checked to determine if the user might be a potential follower.
- 3.
An alternative TF/IDF representation is possible, where the IDF is computed on U and the TF may reflect the level of interaction between user and followed account.
- 4.
This assumption corresponds to a collaborative filtering hypothesis for the related but different task of providing recommendations to the potential followers.
- 5.
- 6.
- 7.
Demonstration video at https://pokedem.futuro.media/.
- 8.
- 9.
Examples of posted charts: https://bit.ly/2LXz1kK, https://bit.ly/2PGq4P5, https://bit.ly/2oIIvXS.
- 10.
Note that the growth from 100 to 1000 followers occurred in different periods for each account. As the growth rate depends on the number of followers previously accumulated, a comparison is possible only for the same growth range.
- 11.
- 12.
References
Armentano, M.G., Godoy, D., Amandi, A.: Topology-based recommendation of users in micro-blogging communities. J. Comput. Sci. Tech. 27(3), 624–634 (2012)
Armentano, M., Godoy, D., Amandi, A.: Followee recommendation based on text analysis of micro-blogging activity. Inf. Syst. 38(8), 1116–1127 (2013)
Barbieri, N., Bonchi, F., Manco, G.: Who to follow and why: link prediction with explanations. In: Proceedings of KDD, pp. 1266–1275 (2014)
Barbu, O.: Advertising, microtargeting and social media. Procedia - Soc. Behav. Sci. 163, 44–49 (2014)
Chaoji, V., Ranu, S., Rastogi, R., Bhatt, R.: Recommendations to boost content spread in social networks. In: Proceedings of WWW, pp. 529–538 (2012)
Corcoglioniti, F., Giuliano, C., Nechaev, Y., Zanoli, R.: Pokedem: an automatic social media management application. In: Proceedings of RecSys, pp. 358–359 (2017)
Corcoglioniti, F., Palmero Aprosio, A., Nechaev, Y., Giuliano, C.: MicroNeel: Combining NLP tools to perform named entity detection and linking on microposts. In: Proceedings of EVALITA (2016)
Cresci, S., Di Pietro, R., Petrocchi, M., Spognardi, A., Tesconi, M.: Fame for sale: efficient detection of fake Twitter followers. Decis. Support Syst. 80, 56–71 (2015)
Gupta, P., Goel, A., Lin, J., Sharma, A., Wang, D., Zadeh, R.: WTF: The who to follow service at Twitter. In: Proceedings of WWW, pp. 505–514 (2013)
Hannon, J., Bennett, M., Smyth, B.: Recommending Twitter users to follow using content and collaborative filtering approaches. In: Proceedings of RecSys, pp. 199–206 (2010)
Kim, Y., Shim, K.: TWILITE: a recommendation system for Twitter using a probabilistic model based on latent Dirichlet allocation. Inf. Syst. 42, 59–77 (2014)
Kywe, S.M., Lim, E.P., Zhu, F.: A survey of recommender systems in Twitter. In: Proceedings of SocInfo, pp. 420–433 (2012)
Lops, P., de Gemmis, M., Semeraro, G.: Content-based recommender systems: state of the art and trends. In: Ricci, F., Rokach, L., Shapira, B., Kantor, P.B. (eds.) Recommender Systems Handbook, pp. 73–105. Springer, Boston (2011). https://doi.org/10.1007/978-0-387-85820-3_3
Nechaev, Y., Corcoglioniti, F., Giuliano, C.: Concealing interests of passive users in social media. In: Proceedings of Re-coding Black Mirror ISWC Workshop (2017)
Nechaev, Y., Corcoglioniti, F., Giuliano, C.: Linking knowledge bases to social media profiles. In: Proceedings of SAC, pp. 145–150 (2017)
Nechaev, Y., Corcoglioniti, F., Giuliano, C.: SocialLink: linking DBpedia entities to corresponding Twitter accounts. In: Proceedings of ISWC, pp. 165–174 (2017)
Nechaev, Y., Corcoglioniti, F., Giuliano, C.: SocialLink: exploiting graph embeddings to link DBpedia entities to Twitter profiles. Prog. Artif. Intell. (2018, to appear)
Nechaev, Y., Corcoglioniti, F., Giuliano, C.: Type prediction combining linked open data and social media. In: Proceedings of CIKM (2018, to appear)
Paniagua, J., Sapena, J.: Business performance and social media: love or hate? Bus. Horiz. 57(6), 719–728 (2014)
Piao, G., Breslin, J.G.: Inferring user interests in microblogging social networks: a survey. User Model. User-Adapt. Interact. 28, 277–329 (2018)
Tommasel, A., Corbellini, A., Godoy, D., Schiaffino, S.: Personality-aware followee recommendation algorithms: an empirical analysis. Eng. Appl. Artif. Intell. 51, 24–36 (2016)
Yuan, G., Murukannaiah, P.K., Zhang, Z., Singh, M.P.: Exploiting sentiment homophily for link prediction. In: Proceedings of RecSys, pp. 17–24 (2014)
Zhao, G., Lee, M.L., Hsu, W., Chen, W., Hu, H.: Community-based user recommendation in uni-directional social networks. In: Proceedings of CIKM, pp. 189–198 (2013)
Acknowledgments
This work was partially supported by the EC Commission through EIT Digital’s High Impact Initiative Street Smart Retail (HII SSR).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Corcoglioniti, F., Nechaev, Y., Giuliano, C., Zanoli, R. (2018). Twitter User Recommendation for Gaining Followers. In: Ghidini, C., Magnini, B., Passerini, A., Traverso, P. (eds) AI*IA 2018 – Advances in Artificial Intelligence. AI*IA 2018. Lecture Notes in Computer Science(), vol 11298. Springer, Cham. https://doi.org/10.1007/978-3-030-03840-3_40
Download citation
DOI: https://doi.org/10.1007/978-3-030-03840-3_40
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-03839-7
Online ISBN: 978-3-030-03840-3
eBook Packages: Computer ScienceComputer Science (R0)