Personalized Web Search
Download as PPTX, PDF5 likes5,565 views
The document discusses the concept of personalized web search (PWS). It notes that generic web search engines cannot identify different user needs, so PWS was introduced to personalize search results. Various techniques for PWS are discussed, including user profiling using demographic and interest data, hyperlink analysis, and community-based or location-based approaches. Maintaining accurate user profiles that respect privacy is also addressed. Potential applications and limitations of PWS are mentioned.
Personalized Web Search
- 2. The significance of effective and efficient information retrieval. Usage of Internet and web technologies. Advent of WWW and web search engines.
- 3. Generic web search engine cannot identify the different needs of different customers. The search results should be personalized to address this issue. Personalized Web Search concept was introduced.
- 4. The importance of the concept of Personalized Web Search. Personal interest to study on this technological area.
- 5. Ability to identify the different needs of different people who issue the same text query for web search. e.g. ‘Matrix’, ‘apple’ Yahoo used this concept in 1998 80 % of the users prefer to use personalized web search engines.
- 6. User profiling Client side implementation Server side implementation Content analysis Hyperlink analysis Community based PWS User location based PWS
- 7. A separate user profile should be maintained for each user. User profile consists with technical, demographical and geographical informations of users. Previously visited pages, total visit time, number of visits , used links, age, gender, education, IP addresses and bookmarks etc.
- 8. Search engine has to maintain user profiles by using its resources. Engine can use its all resources to optimize the search results. Allocate a huge amount of memory and computing processes to maintain millions of user profiles.
- 9. Users /clients are the responsible parties for maintaining their user profiles. An installed software/plugin should be used to facilitate. Violation of privacy and security can be preserved as much as possible. Cost of storage & computing processes are distributing among users. Limitation of network bandwidth.
- 10. This is under User profiling technique Check the similarity between web pages and user profile details. User interested topics and title or content of the web pages are much concerned.
- 11. Most of the leading search engine use this method Crawling and ranking concepts PageRank and Biased PageRank approaches
- 12. Avoid the handling of separate user profile for each user. Search engine has to find the users who have similar kinds of interests. Efficient identification increases the productivity of the collaborative web search.
- 13. Groupization Give higher weights to pages that are relevant to more members of the group. Hit-Highlighting Users’ keywords appeared within the title, snippet or URL of each page are emphasized.
- 14. Some users search things in a particular area. (by implicit local intent queries) E.g. Italian restaurants in San Francisco It uses IP addresses mapping for identifying city name, Zip code Issues Some names are ambiguous like Oakland Some have distinct meanings (Mountain view)
- 15. Google Custom search engine Alpha search engine Google web history tool iGoogle & My Yahoo! Yoople! Collaborative web search engine
- 16. This is an area, the concept can be used in a practical manner. Identify the people in destroyed location by analyzing user profiles. Alert people, before the disaster happens by analyzing scientific stuffs.
- 17. PWS is more time and effort consumable than generic web search. Violation of privacy Ethical and security issues Users’ needs are not static. Handling more complex user profiles.
- 18. How to overcome the limitations. Usage of location based PWS by covering more geographical area. Query expansions Enhance PWS by using the commonsense and folksonomy.
- 19. The concept of PWS is an evolving study area. Adding more techniques frequently. PWS can use for managing the real world scenario like disasters etc.
Editor's Notes
- #8: User profile consists with technical,demographic and geographic information of users.