Random Walks on Context-Aware Relation Graphs for Ranking Social Tags
نویسندگان
چکیده
Social tagging provides an efficient way to manage online resources. In order to collect more social tags, many research efforts aim to automatically suggest tags to help users annotate tags. Many content-based methods assume tags are independent and suggest tags one by one independently. Although it makes suggestion easier, the independence assumption does not confirm to reality, and the suggested tags are usually inconsistent and incoherent with each other. To address this problem, we propose to model contextaware relations of tags for suggestion: (1) By regarding resource content as context of tags, we propose Tag Context Model to identify specific context words in resource content for tags. (2) Given a new resource, we build a context-aware relation graph of candidate tags, and propose a random walk algorithm to rank tags for suggestion. Experiment results demonstrate our method outperforms other state-of-the-art methods. TITLE AND ABSTRACT IN CHINESE 3þe©a 'X ä¥Åir?1 ¬I\üS ¬I\ ́«k +n3℄ a"Ǒ 1⁄4 õ ¬I\§NõïÄ åugÄíI\5Ï<I5"éõÄuSN { ǑI\m ́Õá §Ïd á/ízI\"Ǒ,ù4í{{ü§ ́Õá5b ¿ØÎÜý¢1§l í I\pm3ØÚØÆN"Ǒ )ûù ̄K§·JÑéI\ m þe©a 'X?1ï μ£1¤·ò℄ SNǑI\ þe©§ÏLI\þ e© .lSN¥uyI\ A1⁄2þe©"£2¤ 1⁄2# ℄ §·ïáÿÀI \m þe©'Xã§ÏLÅir{éI\üS¿í"¢ (Jy2· { `u®k{"
منابع مشابه
Random Walks in Social Networks and their Applications: A Survey
A wide variety of interesting real world applications, e.g. friend suggestion in social networks, keyword search in databases, web-spam detection etc. can be framed as ranking entities in a graph. In order to obtain ranking we need a graph-theoretic measure of similarity. Ideally this should capture the information hidden in the graph structure. For example, two entities are similar, if there a...
متن کاملTractable Algorithms for Proximity Search on Large Graphs
Identifying the nearest neighbors of a node in a graph is a key ingredient in a diverse set of ranking problems, e.g. friend suggestion in social networks, keyword search in databases, web-spam detection etc. For finding these “near” neighbors, we need graph theoretic measures of similarity or proximity. Most popular graph-based similarity measures, e.g. length of shortest path, the number of c...
متن کاملSome remarks on the elliptic Harnack inequality
In this note we give three short results concerning the elliptic Harnack inequality (EHI), in the context of random walks on graphs. The first is that the EHI implies polynomial growth of the number of points in balls, and the second that the EHI is equivalent to an annulus type Harnack inequality for Green’s functions. The third result uses the lamplighter group to give a counterexample concer...
متن کاملRandom Walks in Hypergraph
Random walks on graphs have been extensively used for a variety of graph-based problems such as ranking vertices, predicting links, recommendations, and clustering. However, many complex problems mandate a high-order graph representation to accurately capture the relationship structure inherent in them. Hypergraphs are particularly useful for such models due to the density of information stored...
متن کاملThe Einstein relation for random walks on graphs
This paper investigates the Einstein relation; the connection between the volume growth, the resistance growth and the expected time a random walk needs to leave a ball on a weighted graph. The Einstein relation is proved under different set of conditions. In the simplest case it is shown under the volume doubling and time comparison principles. This and the other set of conditions provide the ...
متن کامل