報(bào)告題目：The effect of user generated photos on review helpfulnessprediction: An analytical approach

報(bào)告人：樊衛(wèi)國(guó)

報(bào)告時(shí)間：2018年6月11日（周一）下午3:00-4:30

報(bào)告地點(diǎn)：bwin必贏唯一官網(wǎng)313

講座摘要：Online reviews have been extensively studied in the hospitality andtourism literature. However, while user-provided photos embedded in onlinereviews accumulate in large quantities, their informational value has not beenwell understood likely due to technical challenges. The goal of this study isto introduce deep learning for computer vision to understand information valueof online hotel reviews. Using a dataset collected from two social media sites,we compared deep learning models with other machine learning techniques toexamine the effect of user-provided photos on review helpfulness. Findings showthat deep learning models were more useful in predicting review helpfulnessthan other models. While user-provided photos alone did not have the same impactas review texts, combining review texts and user-provided photos produced thehighest performance. Implications for the applications of deep learningtechnologies in hospitality and tourism research, as well as limitations anddirections for future research, are discussed.

報(bào)告人簡(jiǎn)介：樊衛(wèi)國(guó)教授為美國(guó)愛(ài)荷華大學(xué)Tippie商學(xué)院Tippie商業(yè)分析主席，一直著力于數(shù)據(jù)挖掘，尤其是文本挖掘方向的科學(xué)研究。是國(guó)際上第一個(gè)利用遺傳規(guī)劃算法進(jìn)行搜索引擎排序函數(shù)發(fā)現(xiàn)與優(yōu)化的人，并在此專題上發(fā)表大量具有國(guó)際先進(jìn)水平的文章。同時(shí)在基于互聯(lián)網(wǎng)的網(wǎng)上搜索、自動(dòng)問(wèn)題回答、文本歸納，網(wǎng)上知識(shí)社區(qū)，電子商務(wù)競(jìng)爭(zhēng)模式等方面都頗有建樹。所發(fā)表的100多篇國(guó)際論文已有近2000次索引?，F(xiàn)擔(dān)任Management Science(MS)，Management Information System Quarterly (MISQ)，IEEE Transactions on Evolutionary Computation(TEC)，IEEETransactions on Knowledge and Data Engineering (TKDE)，ACM Transactions on Information Systems (TOIS)等期刊審稿人，以及IEEE Technical Committee on Digital Libraries成員，Journal of Database Management編委會(huì)成員，近20個(gè)國(guó)際會(huì)議評(píng)審委員。