Combating spam in tagging systems: An evaluation
ACM Transactions on the Web (TWEB)
| Hi-index | 0.00 |
The Web is becoming a regular source of information for health information (HI) seekers. In the United States for example, several studies report that more than 80% of Internet users routinely utilize the Web to obtain medical and health information. While Google searching is by far the default point of access for HI seekers, using expert Web sites such as Medline+ and WebMD is also another alternative for users that are looking for quality content. In addition, health expert Web sites are supported with several advanced querying such as browsing using categories organized either as a hierarchy or in an A-Z search list. This mode of searching supported by expert generated taxonomies or metadata is particularly suitable for discovery purposes. In parallel social Web sites such as Delicious and StumbleUpon are providing Web users with a new opportunity to participate in the process of annotating Web resources. This is particularly true for health data as supported by the large amount of health resources annotated by Delicious for example. In this paper we investigate the relationship between expert health metadata and user annotations where the objective is to determine how they can be combined to provide HI seekers with better means to finding their HI needs. The results of the study show that while there is noticeable overlapping between these two types of data, user tags are clearly a new type of metadata that will play an important role in supporting HI seekers searching/browsing health resources.