A Review of Ontology-Based Tag Recommendation Approaches

Abstract

Tag recommender schemes suggest related tags for an untagged resource and better tag suggestions to tagged resources. Tagging is very important if the user identifies the tag that is more precise to use in searching interesting blogs. There is no clear information regarding the meaning of each tag in a tagging process. An user can use various tags for the same content, and he can also use new tags for an item in a blog. When the user selects tags, the resultant metadata may comprise homonyms and synonyms. This may cause an improper relationship among items and ineffective searches for topic information. The collaborative tag recommendation allows a set of freely selected text keywords as tags assigned by users. These tags are imprecise, irrelevant, and misleading because there is no control over the tag assignment. It does not follow any formal guidelines to assist tag generation, and tags are assigned to resources based on the knowledge of the users. This causes misspelled tags, multiple tags with the same meaning, bad word encoding, and personalized words without common meaning. This problem leads to miscategorization of items, irrelevant search results, wrong prediction, and their recommendations. Tag relevancy can be judged only by a specific user. These aspects could provide new challenges and opportunities to its tag recommendation problem. This paper reviews the challenges to meet the tag recommendation problem. A brief comparison between existing works is presented, which we can identify and point out the novel research directions. The overall performance of our ontology‐based recommender systems is favorably compared to other systems in the literature.

Data mining‐based tag recommendation system: an overview

“…Activation begins with certain value that is received by initial nodes and the level of activation is computed by assigning activation. Primary nodes are added in a priority queue arranged with downward activation …”

“…Primary nodes are added in a priority queue arranged with downward activation. 43 On performing the initialization, the highest weighted node is removed from the priority queue and the node activation spreads to all the nearest nodes. These nearest nodes are then added into the queue if they are not labeled as processed.…”

Explainable decision support through the learning and visualization of preferences from a formal ontology of antibiotic treatments

“…when the feature values are not defined for single instances but for classes of instances), a frequent situation in the medical domain. Ontologies have been shown to enhance the performance of preference-based recommendation systems [15,16,17]. Their use has also been proposed for the structuring of preference models [18].…”

How to measure information similarity in online social networks: A case study of Citeulike

“…The idea to focus on the number of common items between users without any consideration of content and metadata is the most close analogy to the current rating-based approach in collaborative filtering and was first to be explored [47]. However, it has been also argued that content-oriented similarity approaches are more meaningful for social systems where the volume of socially shared content (e.g., tweets, photos or bookmarks) is much larger than the volume of items (e.g., movies or books) in traditional recommender systems [41]. As a result of this large information flow, like-minded users tend to have a much lower chance to share or adopt exactly same items because it is infeasible to look through all available information [13].…”