【正文】 remendation (Adomavicius and Tuzhilin 2005). Contentbased remendation is derived from Information Retrieval. A contentbased remendation algorithm identifies and extracts features of items and user and then builds a matching model for them. Remendations are made based on parison of user’s preference and item’s features. On the other hand, the main idea of collaborative filtering is grouping likeminded users together. These systems are also called cliquebased systems. It is assumed that users who had similar choices before will make the same selection in the future. Collaborative remender systems give users suggestion by observing the neighbor of the user. Hybrid remendation mechanisms attempt to deal with some of limitation and overe drawbacks of pure contentbased approach and pure collaborative approach by bining the two approaches.There are several drawbacks when applying existing remendation algorithms to elearning environments directly:Since the learning process is repeatable and periodic, there are some intrinsic orders for learning material in users’ learning processes that can present material access patterns. This information can reflect the learner’s latent preference. But, most of existing remendation systems don’t use this information. To implement a sequential pattern based remendation, the new algorithms are presented in this research. Some of traditional remendation algorithms only use learners’ rating for remendation and don’t consider attributes of learners and learning materials. To model multipreference of learner this research takes into account multidimensionalattribute of materials and learners’ rating matrix in the unified model. The learners’ preferences will be changing dynamically. Therefore, to make good remendation in time when learners’ current interests are changing, a remendation algorithm must trace learner behaviour to propose dynamic remendation. Thus, this research implements a dynamic approach for producing remendations in the multidimensional attributebased CF. According to the described drawbacks, this paper proposes a new material remender system framework and relevant remendation algorithms for elearning environments. First, in the multidimensional attributebased CF remendation approach, to reflect learner’s plete spectrum of interests, Leaner Preference Tree (LPT) is introduced to consider multidimensionalattributes of materials, learner’s rating simultaneously. Truly, Leaner Preference Tree is built based on target learner’s historical access records and multidimensionalattributes of materials. Then, a new similarity measure that can take into account the information of LPTs for calculating similarity between learners is introduced. In the sequential pattern based remendation approach, to discover the latent patterns of accessed materials and give remendation, the weighted association rules (Apriori algorithm) and PrefixSpan algorithm are implemented. The results of two approaches are bined to create final remendations.The rest of this paper is　 Literature survey，In Literature survey section, the previous related works on elearning material remender systems are discussed.Learning materials have grew either offline or online in educational organizations. So, it is difficult for learners to discover the most appropriate materials according to keyword searching methods. The creation of the technology for personalized lifelong learning has been recognized as a Grand Challenge Problem by peak research bodies (Kay 2008). Therefore, remender systems have been used for elearning environments to remend useful materials to users. The first remender system was developed in the mid of 1990s (Felfernig et al. 2007). Many remendation systems in various fields such as movies, music, news, merce and medicine have been developed but few in education field (Drachsler et al. 2007). The Overview of the remendation strategies and techniques with their usefulness for material remendation have been presented in Table 1. We briefly survey some of important works and explain the drawbacks of them that can be addressed by our proposed approach.Content based filtering This technique suggests items similar to the ones that each user liked in the past taking into account the object content analysis that the user has evaluated in the past (Lops et al. 2011). As an example for elearning application, Khribi et al. ( 2009) used learners39。老師嚴謹求實的學(xué)術(shù)態(tài)度、不斷創(chuàng)新的研究作風(fēng)和誨人不倦的導(dǎo)師風(fēng)范讓我受益匪淺，為師和為人之道也是我今后學(xué)習(xí)和工作的榜樣。 Koper, R. (2008). Personal remender systems for learners in lifelong learning: requirements, techniques and model. International Journal of Learning Technology, 3(4), 404–423.[7]Felfernig, A., Friedrich G., SchmidtThieme, L., (2007). Introduction to the IEEE Intelligent Systems Special Issue: Remender Systems, 22(3) 18–21致謝辭時光飛逝，歲月如梭。在此感謝我的畢業(yè)設(shè)計指導(dǎo)老師包云霞，謝謝您在畢業(yè)設(shè)計期間對我的指導(dǎo)。最后一周是對以前的資料和設(shè)計思路進行整理。通過這樣的一個自己從開始到結(jié)束全程自己參與的設(shè)計來說對知識的了解和掌握是純理論的學(xué)習(xí)遠遠達不到的效果。在這個過程中，老師的的精心指導(dǎo)、與同學(xué)的交流、在圖書館查找資料、系統(tǒng)的完善，每一個過程都是對自己能力的一次檢驗和充實。示例代碼：在持久層中，則依賴于Hibernate的對象化映射和數(shù)據(jù)庫交互，處理DAO組件請求的數(shù)據(jù)，并返回處理結(jié)果。　　　基于OJ數(shù)據(jù)的習(xí)題個性化推薦系統(tǒng)采用傳統(tǒng)的登陸注冊頁面，系統(tǒng)新用戶首次登陸系統(tǒng)，后臺會進行認證（非系統(tǒng)內(nèi)記錄學(xué)生不能進行注冊登錄），參見下圖:當(dāng)系統(tǒng)內(nèi)預(yù)設(shè)的用戶首次注冊激活賬號，此時用戶的賬號處于激活狀態(tài)，后臺認證該用戶，并且返回主頁，左上角每一個系統(tǒng)用戶都可以對個人信息進行管理：　　　為了方便習(xí)題推薦系統(tǒng)內(nèi)部用戶的管理，只有系統(tǒng)管理員可以對所有用戶進行增加，刪除，修改，查詢，支持線上利用模板導(dǎo)入、導(dǎo)出用戶到Excel。（由于目前源數(shù)據(jù)中沒有將習(xí)題進行難度分類，所以該字段的初始化為3）； t答題時間：限制為一個小時，分為三個時間段（10分鐘以內(nèi)，30分鐘以內(nèi)，60分鐘以內(nèi)，分為1，2，3個階段）（源數(shù)據(jù)中沒有學(xué)生答題時間的記錄，初始化為2）；處理結(jié)果：初始化系統(tǒng)中的三個類，Student,Content,SCMap。：Web顯示層數(shù)據(jù)訪問層數(shù)據(jù)庫連接數(shù)據(jù)庫MySQLWeb 顯示層即為JSP頁面層，為用戶提供應(yīng)用程序的訪問，本論文中的系統(tǒng)以Web頁面的形式實現(xiàn)。通過比例因子對Contents Contents2進行加權(quán)得到在基于記憶的過濾下的候選題ContentA。（由于目前源數(shù)據(jù)中沒有將習(xí)題進行難度分類，所以該字段的初始化為3）； t答題時間：限制為一個小時，分為三個時間段（10分鐘以內(nèi)，30分鐘以內(nèi)，60分鐘以內(nèi)，分為1，2，3個階段）（源數(shù)據(jù)中沒有學(xué)生答題時間的記錄，初始化為2）?；谝粋€新用戶隸屬每個用戶群的概率和用戶群對推薦習(xí)題的概率，就可以預(yù)測新用戶對某習(xí)題的感興趣程度并進行推薦。在該層，先對數(shù)據(jù)進行預(yù)處理，再將處理后的數(shù)據(jù)集置于學(xué)習(xí)方案中，進行相應(yīng)的挖掘任務(wù)。WEKA 數(shù)據(jù)挖掘平臺完整、實用、高水準地實現(xiàn)了許多流行的學(xué)習(xí)方案，這些方案能夠直接運用于一些實際的數(shù)據(jù)挖掘或研究領(lǐng)域。并且存在熱門項目容易被過度推薦的問題。關(guān)聯(lián)規(guī)則挖掘中有兩個主要的概念，支持