本文選題：旅游網(wǎng)絡廣告 + 個性化推送　； 參考：《湖南工業(yè)大學》2017年碩士論文

【摘要】：旅游網(wǎng)絡廣告是加快旅游業(yè)信息化進程的關鍵手段之一,而旅游個性化信息服務是提升網(wǎng)絡廣告效果的重要方法。然而,愈來愈多的旅游產(chǎn)品、旅游方案的積累和在線旅游人數(shù)的暴增,不僅造成在線游客需要花費較長時間尋找滿足自己需求的內(nèi)容,還導致旅游網(wǎng)站信息量過載,造成在線游客時間成本增加和網(wǎng)站廣告推薦效率下降。鑒于此,本文依據(jù)艾賓浩斯遺忘規(guī)律和馬太效應現(xiàn)象,提出相應的廣告?zhèn)�性化推薦優(yōu)化技術,旨在提高推薦的精準度。首先在游客興趣變化遵循艾賓浩斯遺忘規(guī)律的基礎上,將興趣遺忘函數(shù)融入?yún)f(xié)同過濾算法中,賦予游客評分時間權重,以此削弱歷史評分的權值加強當前評分的重要性。然后分析了推薦系統(tǒng)中馬太效應對游客興趣預測的影響,將旅游產(chǎn)品的流行度引入?yún)f(xié)同過濾算法,加大對熱門產(chǎn)品的懲罰值,降低流行度對游客興趣相似度的影響。最后,從游客歷史評分信息和產(chǎn)品隱含信息對游客興趣預測的影響出發(fā),結合艾賓浩斯遺忘規(guī)律和馬太效應現(xiàn)象,建立了基于遺忘函數(shù)和旅游產(chǎn)品流行度的個性化旅游網(wǎng)絡廣告推薦模型,并提出三種改進流行度的方法,深入剖析流行度對推薦的影響,并結合相關案例進行數(shù)據(jù)仿真。研究結論表明,同時考慮遺忘函數(shù)和產(chǎn)品流行度,及改進的流行度模型比單一角度優(yōu)化的模型預測精準度高,且單一角度優(yōu)化的預測模型推薦精準度均高于傳統(tǒng)協(xié)同過濾算法。優(yōu)化的廣告推薦方法不僅消除在線游客興趣變化和產(chǎn)品流行度對推薦精準度的影響,同時緩解了網(wǎng)站信息量過載和降低了游客瀏覽的時間成本,為網(wǎng)站的精準化廣告推薦提供一定的方法和手段,拓寬了旅游個性化推薦的研究思路。
[Abstract]:Tourism online advertising is one of the key means to speed up the information process of tourism, and tourism personalized information service is an important way to improve the effect of Internet advertising. However, the increasing number of tourism products, the accumulation of tourism plans and the increasing number of online tourists will not only take time for online tourists to find a long time to meet themselves. The content of the demand also leads to the overload of the tourist website information, which causes the increase of the time cost of the online tourists and the decline of the website advertising recommendation efficiency. In view of this, this paper puts forward the corresponding personalized recommendation optimization technology based on the Ebbinghaus's forgetting law and the Matthew effect, aiming at improving the accuracy of the recommended. First, the change of the interest of tourists. On the basis of Ebbinghaus's forgetting law, the interest forgetting function is integrated into the collaborative filtering algorithm, and the weight of tourists' scoring time is given to weaken the importance of the historical score to strengthen the importance of the current score. Then, the influence of the Matthew effect on the tourist interest pretest is analyzed, and the popularity of the tourist product is introduced into the synergy. The filtering algorithm increases the penalty value for popular products and reduces the influence of popularity on the similarity of tourists' interest. Finally, based on the influence of tourist history score information and product implied information on the tourist interest prediction, combining the Ebbinghaus forgetting law and the Matthew effect, it builds a forgetting function and the popularity of tourism products. The recommendation model of sexual tourism network advertising, and three ways to improve the popularity, in-depth analysis of the impact of popularity on recommendation, and the combination of related cases for data simulation. The conclusion shows that the forgetting function and product popularity are considered at the same time, and the improved popularity model is more accurate than the single angle optimization model. The single angle optimization prediction model recommends that accuracy is higher than the traditional collaborative filtering algorithm. The optimized advertising recommendation method not only eliminates the influence of online tourist interest changes and product popularity on the recommended accuracy, but also relieves the amount of information overload and reduces the time cost of visitors' browsing, and promotes the accurate advertising of the website. Recommendation provides certain methods and means to broaden the research idea of personalized tourism recommendation.
【學位授予單位】：湖南工業(yè)大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：F592;F713.8

【參考文獻】

相關期刊論文 前10條

1 李霞;尹川東;袁云;;旅游路線個性化推薦算法比較分析[J];計算機技術與發(fā)展;2016年09期

2 方瀟;劉曉寒;柴永平;周文曼;;一種基于協(xié)同過濾的旅游行程推薦算法[J];地理空間信息;2016年07期

3 李倩;許偉;蔣洪迅;;考慮用戶口碑的旅游計劃個性化推薦方法研究[J];管理評論;2016年06期

4 吳清霞;周婭;文締堯;賀正紅;;基于用戶興趣和興趣點流行度的個性化旅游路線推薦[J];計算機應用;2016年06期

5 牛朋飛;;Web數(shù)據(jù)挖掘技術在電子商務中的應用[J];計算機與網(wǎng)絡;2015年07期

6 王偉;王洪偉;孟園;;協(xié)同過濾推薦算法研究:考慮在線評論情感傾向[J];系統(tǒng)工程理論與實踐;2014年12期

7 張磊;;基于遺忘曲線的協(xié)同過濾研究[J];電腦知識與技術;2014年12期

8 于海平;林曉麗;劉會超;;基于數(shù)據(jù)挖掘的移動廣告?zhèn)�性化推薦研究[J];計算機技術與發(fā)展;2014年08期

9 古凌嵐;;基于GEP-RBF的協(xié)同過濾數(shù)據(jù)稀疏性問題研究[J];計算機與數(shù)字工程;2013年09期

10 朱國瑋;周利;;基于遺忘函數(shù)和領域最近鄰的混合推薦研究[J];管理科學學報;2012年05期

相關博士學位論文 前2條

1 趙向宇;Top-N協(xié)同過濾推薦技術研究[D];北京理工大學;2014年

2 史e，

本文編號：2010958