Open AccessExploiting contextual information to improve call prediction
Author(s) -
Mehk Fatima,
Aimal Rextin,
Shamaila Hayat,
Mehwish Nasim
Publication year - 2019
Publication title -
plos one
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.99
H-Index - 332
ISSN - 1932-6203
DOI - 10.1371/journal.pone.0223780
Subject(s) - computer science , phone , dimension (graph theory) , mobile phone , data mining , machine learning , state (computer science) , artificial intelligence , information retrieval , algorithm , mathematics , telecommunications , philosophy , linguistics , pure mathematics
With the increase in contact list size of mobile phone users, the management and retrieval of contacts has becomes a tedious job. In this study, we analysed some important dimensions that can effectively contribute in predicting which contact a user is going to call at time t . We improved a state of the art algorithm, that uses frequency and recency by adding temporal information as an additional dimension for predicting future calls. The proposed algorithm performs better in overall analysis, but more significantly there was an improvement in the prediction of top contacts of a user as compared to the base algorithm.