AIJ-200 American International Journal of Contemporary Scientific Research ISSN 2349 – 4425 Query search on assortment websites by GMDH using neural network while not CRAWL to get rid of pretend rank websites Zuber Ansari Department of Information Technology, TIT Excellence, RGPV University, Bhopal – 462001 [email protected] ABSTRACT Internet is an assortment of immense data. In the modern internet era everyone search on internet for some information. To explore the internet Search Engines are primary targets. Everyone uses Search Engine from find a name of website for information. Search Engines have primary responsibility to search query related information and rank them according to relevant information. This vast need of Search Engines makes them an important tool for this challenging task. Search engine directly compare keywords with URI, METADATA, CONTENT, REFBACK LINK, web pages against the user query and a web page which has most matched keyword density, results high rank in Search Engine results. In that case many websites want to show their websites on first page against user query search keywords. To get first page ranking Search Engine Optimization is a method to boost ranking of website. Search Engine Optimization is a technique that by pass the Google search results rules as well as shows as first page rank website to the crawler when crawl web pages of website. In this era many fake ranking websites show on first page with irrelevant to keyword as well as user mistrusted websites. In this project we using a neural network method to filter query on website group or set of group and predict relevant website after that search on that website for query. As well as this proposed system work without crawler. Keywords: SEO, SEF, KEYWORD, URI, METADATA, CONTENT, REFBACK LINK, Crawler, Crawl 1. INTRODUCTION Search Engine plays an important role to extract information from internet in the form of ranked web pages result. Searched results top few results got hits form users and low ranked pages. Web page ranking can be improved with Search Engine Optimization, which helps in analysis of Search Engine rules and develop a web page according to Search Engine ranking rules. SEO is a practice which helps to improve web page’s popularity as well as content, which easily recognized by Search Engine. But judge a web page’s quality and popularity is totally a responsibility of Search Engine. Every Search Engine works with some searching algorithms and strategies. On the behalf of searching techniques Search Engine can be divided into two major categories: Keyword Based Search and Semantic Search. Keyword Based Search focuses on the web pages keywords and directly matched with query keywords. Before starting search on internet Search Engines register every web page with their personal database. Search Engines spider are located over the web and make searching of web pages on web easier. Every Search Engine firstly parse the web pages, ignore all the stop words like „the‟ „is‟ „am‟ „are‟ etc. Every article lies between <page> and </page> tags, where the page id, title, and text is separated by the corresponding tags. [2] [18] Interrogative words are removed from the query and page ranking is performed on the basis of main keywords of the query. Every Search Engine performs following three tasks: A. Inverted Index: After parsing all web pages documents Search Engine Create an inverted index for all documents which speed up information retrieval. Example in below table „computer‟ is a term for inverted index and corresponding „ [1, [0, 5, 15]] ‟ is an index for „laptop‟ time period in inverted index database. Here 1 factor to net web page index and [0, 5, 15] factors to term’s index inside web page. Finally search starts towards inverted indexed database. [12] [21] 11| www.yloop.in AIJ-200 American International Journal of Contemporary Scientific Research ISSN 2349 – 4425 (TERM) (DOCID) Computer [ [1, [0, 5, 15]], [2[0, 7, 17]] Literature [[4, [0, 6, 13]], [5[0, 19]] Spirituality [[7, [0, 9, 23]] Search Engines are mainly three types: Full Text, Meta and Directory Search Engine. Search engine have two major searching techniques used to search web pages on the Internet and store them into Search Engine’s personal database and ranking in SERP. [4] B. Query Analysis: User puts a query to search some type of information on the internet. Question works as a feed to look Engine, and explicit consumer s view for seek. On the idea of query keywords question can be divided into three classes: Navigational, Transactional and Informational. Query evaluation is approaches which refine question with shorten key phrases. Interrogative phrases are removed from the question and web page ranking is performed on the idea of primary key phrases of the question. [4] C. Ranking Web Pages: Finally Search Engine rank a web page with some predefined policy, strategies. Ranking a web page may depend on several factors of web pages like Popularity, contents, structure of web pages etc. After search on personal database Search Engine finally rank web pages according to their content frequency matched with query keywords. [4] On the premise of query all search engines may be categorize as keyword based Search Engine or Semantic search Engine. Both types of Search Engines follow all the upper steps but difference is based on the searching techniques. [4] Keyword Based Search: This search mainly focus on the query keywords and match theses query keywords directly with the web page keywords without any proper query analysis or knowledge. Search mainly points out the title tag, anchor text, page heading tag, Meta tag’s keywords, anchor tag, ALT tag, URL. To optimize a web page in search result keywords density, distribution and keywords selection is more important. Every value in a field considered as a small text document that can be used for the Keyword Based Search. Keyword matching process finds a set of tuple in database, which has matched keywords in user query. With the help of inverted index it is easy to determine matched keywords and their index in a webpage. A web page which has exact query matched keywords and has more strength of query related keywords gets high rank in search result. Because of Search Engine gives priority to keywords so Black Hat techniques. Keyword Based Search Engines are Google, Yahoo, and MSN etc. [15] [17] [19] Semantic Search: This is also called a search with meanings. The Semantic Search determines the content and contextual meaning of query keywords. The Semantic Search Engine interpreted extracting the relevant concept from the sentence. A query also itself determines the user’s view. Interrogative words are used to change the meaning of query like what, why, when etc. The Semantic Search Engines are an alternative to the Keyword Based Search Engines. The distinction of Semantic Search Engines from Conventional Search Engines is that the Semantic Search Engines are which means-based totally. Semantic search is combined with conventional key-word-primarily based retrieval to reap tolerance to understanding base incompleteness. The Semantic Search uses a knowledge base to extend the user query. A user uses shortcut keywords but want more results, so this is the Search Engine’s responsibility to extend the user’s query to find relevant results to user query. Knowledge base is a group of related terms which has similar type of related terms. Any keyword which exists in knowledge base can be a part of user query indirectly. Knowledge base mostly may be in form of database with different related terms collection. Each term has pre assigned weight in a database. Term’s weight further used to calculate the web page’s weight for ranking in search results. The Semantic Search also retains the exact meaning of query and relate with results. [16] [20] 1.1 SEARCH ENGINE OPTIMIZATION SEO is the manner to get your site higher ranking to show on first page in search results of search engines. SEO is a necessary service as the Internet is evolving at a rapid pace and there is greater than before competition for market go halves to the top positions in various search engines to reach. Through research help, it was shown that most people prefer the first ten search outcome as if they traverse the list of search engines and rarely go for results. SEO is one of the most overlooked areas of marketing today. You are having the older generation still trying to advertise on billboards, TV commercials, advertisements or radio. In the meantime the younger age group consumed with the hype of Face book, Twitter and additional sites of social networking. There are lots of SEO techniques with best quality that are able to achieve the desired results. SEO is the best publicity method that is much longer term. You can never climb in the rankings of Google within days. Typically it takes several months before seeing a boost in your position. But on one occasion you can even get to the peak of the search engines and you can even receive a constant stream of hundreds. The objective of this research paper was to prove that implementing search engine optimisation elements that are increasing Website traffic and online visibility .For Website usability seo attributes is essential to improve rankings [14] 12| www.yloop.in AIJ-200 American International Journal of Contemporary Scientific Research ISSN 2349 – 4425 1.2 HOW DOES A SEARCH ENGINE WORK? Search engines perform several activities in order to deliver search results. [14] Crawling - Process of fetching all the web pages linked to a website. This task is performed by software called a crawler or a spider (or Google bot, in case of Google).