發(fā)布時間：2018-08-30 11:57

【摘要】：當今的世界是信息化的世界，隨著互聯(lián)網(wǎng)的廣泛使用，越來越多的不具備數(shù)據(jù)庫查詢知識的人需要訪問在線數(shù)據(jù)庫，由此產(chǎn)生了基于關鍵字的搜索引擎，它是人們在互聯(lián)網(wǎng)上搜索信息的重要方式。數(shù)據(jù)庫中的關鍵字搜索技術(shù)正是在這樣的背景下應運而生，成為數(shù)據(jù)庫、信息檢索、互聯(lián)網(wǎng)等領域的學者共同關注的研究熱點。 關系數(shù)據(jù)庫中的關鍵字查詢問題使得用戶不需要具備數(shù)據(jù)庫相關知識，也不需要掌握復雜的SQL語法就可以像使用互聯(lián)網(wǎng)搜索引擎一樣使用關鍵字來對數(shù)據(jù)進行查詢，因此受到人們的廣泛關注，許多方法和原型被提出。當前流行的關系數(shù)據(jù)庫關鍵字查詢技術(shù)存在較慢的查詢時間或者不準確的查詢結(jié)果。針對這兩個問題，本文提出了一種基于用戶反饋的查詢方法，，該方法從以下兩個方面進行研究： 1.提供給用戶一個動態(tài)表單，使用戶能夠參與查詢過程。根據(jù)用戶輸入的關鍵字動態(tài)的生成一個在數(shù)據(jù)庫中包含關鍵字的屬性所組成的Form表單提供給用戶，用戶在Form中進行選擇并提交，最后，根據(jù)用戶的選擇和數(shù)據(jù)庫模式圖進行連接算法并執(zhí)行SQL獲得最終結(jié)果，實驗證明該方法縮短了查詢時間并且提高了查詢的準確性。 2.在查詢結(jié)果中滿足條件的元組連接樹可能很多，但是用戶通常只對與關鍵字最相關的結(jié)果感興趣。所以需要對得到的結(jié)果集(元組連接樹)使用排序函數(shù)進行排序，把最相關的結(jié)果排在結(jié)果集的前面返回給用戶。排序函數(shù)是以評分函數(shù)為基礎的，評分函數(shù)為每個元組連接樹賦予一個分數(shù)，然后，排序函數(shù)根據(jù)這個分數(shù)按照降序?qū)υM連接樹進行排列。本文提出了對返回結(jié)果集的一種排序策略。以查詢序列與結(jié)果元組樹之間的相似值作為排序依據(jù)，并分析與重新定義了相關影響因子的標準化函數(shù)表達式。通過在簡單數(shù)據(jù)庫上的分析驗證了該改進是合理的。
[Abstract]:Today's world is the world of information, with the wide use of the Internet, more and more people who do not have knowledge of database query need to access online databases, resulting in a keyword-based search engine. It is an important way for people to search for information on the Internet. Keyword search technology in database emerges as the times require under this background and has become a common research hotspot in the fields of database, information retrieval, Internet and so on. The key query problem in relational database makes users do not need to have knowledge of database, and do not need to master complex SQL syntax to use keywords to query data just like Internet search engine. As a result, many methods and prototypes have been proposed. At present, the popular key query technology of relational database has slow query time or inaccurate query results. Aiming at these two problems, this paper proposes a query method based on user feedback, which is studied from the following two aspects: 1. Provides the user with a dynamic form to enable the user to participate in the query process. Dynamically generate a Form form composed of attributes that contain keywords in the database according to the keywords entered by the user. The user selects and submits the form in the Form. According to the user's choice and database schema diagram, the algorithm is connected and SQL is executed to obtain the final result. The experiment shows that the method shortens the query time and improves the accuracy of query. 2. There may be many tuple join trees that satisfy the criteria in query results, but users are usually only interested in the results that are most relevant to the keyword. So it is necessary to sort the result set (tuple join tree) using sort function and return the most relevant result to the user before the result set. The sorting function is based on the scoring function, which assigns a score to each tuple join tree, and then sorts the tuple join tree in descending order according to the score. In this paper, a sorting strategy for returning result sets is proposed. Based on the similarity between the query sequence and the result tuple tree, the normalized function expressions of the related factors are analyzed and redefined. The analysis on the simple database proves that the improvement is reasonable.
【學位授予單位】：黑龍江大學
【學位級別】：碩士
【學位授予年份】：2012
【分類號】：TP311.13

【參考文獻】

相關期刊論文 前3條

1 王珊;張俊;彭朝暉;戰(zhàn)疆;杜小勇;;基于本體的關系數(shù)據(jù)庫語義檢索[J];計算機科學與探索;2007年01期

2 文繼軍,王珊;SEEKER:基于關鍵詞的關系數(shù)據(jù)庫信息檢索[J];軟件學報;2005年07期

3 彭朝暉;張俊;王珊;;S-CBR:基于數(shù)據(jù)庫模式展現(xiàn)數(shù)據(jù)庫關鍵詞檢索結(jié)果[J];軟件學報;2008年02期

本文編號：2213002