Follow us on Facebook

Header Ads

Problems and Solutions of Web Search Engines


Problems and Solutions of Web Search Engines
ABSTRACT

In order to study the retrieval precision of network information and solve the problems existed in search engines this paper analyzes problems and precision in the information retrieval according to experimental data such as image retrieval with different retrieval keywords, and puts forward a new construction of search engine with intelligentization. It comes to a conclusion that many methods can be used to improve the network retrieval precision effectually. And users need to meet with correct solution to keywords, construction and improvement of knowledge library, reasonable definition of features vectors, information matching and filtering, and increasing the intelligentization for spider, indexer and searcher.

EXISTING SYSTEM

  • Many references only introduce the given key words and the target number of the search engines; they do not carefully analyze how much useful information the target pages contain.
  • In the existing system, the search is made generally, so all the contents are put in one page. So many irrelevant and mismatch content dump the search pages.
  • The existing system does not categorize search made on Content based and location based. This is one of the most user preference.
  • The existing system has many disadvantages. Few are
o       Time Complexity
o       Complex Queries
o       Less User Interactivity



PROPOSED SYSTEM
  • In the proposed system, Search is categorized on two main categorizes viz.: Content based Search and Location Based Search.
  • Content searching linked the ontology shows the possible concept space arising from a user's queries. In this ontology covers more than what the user actually wants.
  • In Location based search, the contents are viewed to the user based on the location updated by the user.
  • The proposed system supports good time complexity, complex queries and User Interactive.

HARDWARE SPECIFICATION
PROCESSOR             :                       PENTIUM 4 CPU 2.40GHZ
RAM                           :                       128 MB
            HARD DISK               :                       40 GB
            KEYBOARD              :                        STANDARD
            MONITOR                  :                       15”

SOFTWARE SPECIFICATION
FRONT END                          :                      C#.NET
BACK END                            :                       SQL SERVER 2000
OPERATING SYSTEM          :                       WINDOWS XP
DOCUMENTATION              :                       MS-OFFICE 2007

MODULES:
·        Profile Registration
·        Content Searching
·        Location Searching
Profile Registration
In this user has to register the user information and it will provide the login for maintaining the information. It also maintains the searched data which should be useful for next searching .it should automatically rank depends upon the user interest upon the particular search. It also re-ranked whenever the searching criteria have been modified. In this user profile contains not only profile information and also search content which helps to search and give immediate results whatever information user needed.


DATA FLOW DIAGRAM FOR PROFILE REGISTERATION


Content Searching
Content searching linked the ontology shows the possible concept space arising from a user's queries. In this ontology covers more than what the user actually wants. When the query is submitted, the data for the query composes of various relevant data. If the user is indeed interested in some specific data means the click through is captured and the clicked data is favored. The content ontology together with the click through serves as the user profile in the personalization process. It will then be transformed into a linear feature vector to rank the search results according to the user's content information preferences.

DATA FLOW DIAGRAM FOR CONTENT SEARCHING


Location Searching
In this module extracting location concepts is different from that for extracting content concepts. First, a document usually embodies only a few location concepts. As a result, very few of them co-occur with the query terms in web- snippets. We extract location concepts from the full documents. Second, due to the small number of location concepts embodied in documents, the similarity and parent-child relationship cannot be accurately derived statistically.


                              DATA FLOW DIAGRAM FOR LOCATION SEARCHING


 

REFERENCE:

Wang Liangshen, Hou Jie, Xie Zaiyu, Wang Xiaochen, Que Caiyue, Li Hui, “Problems and Solutions of Web Search Engines”, IEEE International Conference on Consumer Electronics, Communications and Networks (CECNet), 2011.