Abstract
In the common formulation, the recommendation problem is reduced to the problem of estimating the utilization for the items that have not been seen by a user [1]. Micro-blog recommendation will recommend micro-blogs interest users, mostly those related to the micro-blogs that a user had issued or trending topics. One indispensable step in realizing effective recommendation is to compute short text similarities between micro-blogs. In this paper, we utilize two kinds of approaches, traditional cosine-based approach and WordNet-based semantic approach, to compute similarities between micro-blogs and recommend top related ones to users. We conduct experimental study on the effectiveness of two approaches using a set of evaluation measures. The results show that semantic similarity based approach has relatively higher precision than that of traditional cosine-based method using 548 twitters as dataset. Copyright © 2012 Springer.
| Original language | English |
|---|---|
| Title of host publication | Database systems for advanced applications: 17th International Conference, DASFAA 2012, International Workshops: FlashDB, ITEMS, SNSM, SIM3, DQDI, Busan, South Korea, April 15-18, 2012, Proceedings |
| Editors | Hwanjo YU, Ge YU, Wynne HSU, Yang-Sae MOON, Rainer UNLAND, Jaesoo YOO |
| Publisher | Springer |
| Pages | 115-122 |
| ISBN (Print) | 9783642290220 |
| DOIs | |
| Publication status | Published - 2012 |