Topical affinity in short text microblogs

Abstract Knowledge-based applications like recommender systems in social networks are powered by complex network of social discussions and user connections. Short text microblog platforms like Twitter are powerful in this aspect due to their real-time content dissemination as well as having a complex mesh of user connections. For example, users on Twitter tend to consume certain content to a greater or less extent depending on their interests over time. Quantifying this degree of content consumption in certain topics is an arduous task. This is further compounded by the amount of digital information that such platforms generate at any given time. Formulation of personalized user profiles based on user interests over time and friendship network is thus a problem. Therefore, user profiling based on their interests is important for personalized third-party content recommendations on the platform. In this paper we address this problem by presenting our solution in a two-step process:- (i) Firstly, we compute users’ Degree of Interest (DoI) towards a certain topic based on the overall users’ affinity towards that topic. (ii) Secondly, we affirm this DoI by correlating it to their friendship network. Furthermore, we describe our model for DoI computation and follow-back recommendation system by learning a low-dimensional vector representation of users and their disseminated content. This representation is used to train models for prediction of correct cluster classifications. In our experiments, we use a Twitter dataset to validate our approach by computing degrees of interest for certain test users in three diverse and generic topics. Experimental results show the effectiveness of our approach in the extraction of intra-user interests and better accuracy in follow-back recommendations with diversities in the topics.

[1]  Jeffrey Dean,et al.  Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[2]  Gerhard Nahler,et al.  Pearson Correlation Coefficient , 2020, Definitions.

[3]  Lorena Recalde,et al.  Who is suitable to be followed back when you are a Twitter interested in Politics? , 2017, DG.O.

[4]  M. McPherson,et al.  Birds of a Feather: Homophily in Social Networks , 2001 .

[5]  Michael S. Bernstein,et al.  Short and tweet: experiments on recommending content from information streams , 2010, CHI.

[6]  El Habib Nfaoui,et al.  Using Tweets Embeddings For Hashtag Recommendation in Twitter , 2018 .

[7]  Jan Treur,et al.  An adaptive temporal-causal network model for social networks based on the homophily and more-becomes-more principle , 2019, Neurocomputing.

[8]  Miki Haseyama,et al.  Twitter Followee Recommendation Based on Multimodal FFM Considering Social Relations , 2018, 2018 IEEE 7th Global Conference on Consumer Electronics (GCCE).

[9]  Alípio Mário Jorge,et al.  Identifying topic relevant hashtags in Twitter streams , 2019, Inf. Sci..

[10]  Sergei Vassilvitskii,et al.  k-means++: the advantages of careful seeding , 2007, SODA '07.

[11]  Brian G. Knight,et al.  Homophily, Group Size, and the Diffusion of Political Information in Social Networks: Evidence from Twitter , 2014 .

[12]  Jia Li,et al.  Tweet modeling with LSTM recurrent neural networks for hashtag recommendation , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[13]  Brij B. Gupta,et al.  Hybrid approach for detection of malicious profiles in twitter , 2019, Comput. Electr. Eng..

[14]  Yong Yu,et al.  Collaborative personalized tweet recommendation , 2012, SIGIR '12.

[15]  Peter Brusilovsky,et al.  User Models for Adaptive Hypermedia and Adaptive Educational Systems , 2007, The Adaptive Web.

[16]  Yannis Stavrakas,et al.  Tweet and followee personalized recommendations based on knowledge graphs , 2018, J. Ambient Intell. Humaniz. Comput..

[17]  JawaheerGawesh,et al.  Modeling User Preferences in Recommender Systems , 2014 .

[18]  Cesar H. Comin,et al.  Clustering algorithms: A comparative approach , 2016, PloS one.

[19]  P. Rousseeuw.  Silhouettes: a graphical aid to the interpretation and validation of cluster analysis , 1987 .

[20]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[21]  Qi Gao,et al.  Analyzing user modeling on twitter for personalized news recommendations , 2011, UMAP'11.

[22]  Shuang-Hong Yang,et al.  Large-scale high-precision topic modeling on twitter , 2014, KDD.

[23]  Panagiotis Symeonidis,et al.  Feature-Weighted User Model for Recommender Systems , 2007, User Modeling.

[24]  Ravinder Kumar,et al.  Folksonomy-based user profile enrichment using clustering and community recommended tags in multiple levels , 2018, Neurocomputing.

[25]  Tomas Mikolov,et al.  Enriching Word Vectors with Subword Information , 2016, TACL.

[26]  Yang Liu,et al.  A User Adaptive Model for Followee Recommendation on Twitter , 2016, NLPCC/ICCPOL.

[27]  Zhiting Hu,et al.  Dynamic User Modeling in Social Media Systems , 2015, TOIS.

[28]  M. McHugh.  Interrater reliability: the kappa statistic , 2012, Biochemia medica.

[29]  Jon Atle Gulla,et al.  Real-time social recommendation based on graph embedding and temporal context , 2019, Int. J. Hum. Comput. Stud..

[30]  Chin-Teng Lin,et al.  A review of clustering techniques and developments , 2017, Neurocomputing.

[31]  Juan-Zi Li,et al.  Typicality-Based Collaborative Filtering Recommendation , 2014, IEEE Transactions on Knowledge and Data Engineering.

[32]  Jeffrey Dean,et al.  Distributed Representations of Words and Phrases and their Compositionality , 2013, NIPS.

[33]  Hongfei Yan,et al.  Comparing Twitter and Traditional Media Using Topic Models , 2011, ECIR.

[34]  Jiangtao Ren,et al.  Forum latent Dirichlet allocation for user interest discovery , 2017, Knowl. Based Syst..

[35]  Hui Tian,et al.  A new user similarity model to improve the accuracy of collaborative filtering , 2014, Knowl. Based Syst..

[36]  David Robinson,et al.  Detecting Hate Speech on Twitter Using a Convolution-GRU Based Deep Neural Network , 2018, ESWC.

[37]  Charu C. Aggarwal,et al.  Content-Based Recommender Systems , 2016 .

[38]  Purnima Bholowalia,et al.  EBK-Means: A Clustering Technique based on Elbow Method and K-Means in WSN , 2014 .