Johann Stan's research while affiliated with University of Lyon and other places

Publications (5)

Article
In recent years, online collaborative environments, e.g. social content sites (such as Twitter or Facebook) have significantly changed the way people share information and interact with peers. These platforms have become the primary common environment for people to communicate about their activity and their information needs and to maintain and cre...
Chapter
In this book chapter the authors address two main challenges for building compelling social applications. In the first challenge they focus on the user by addressing the issue of building dynamic interaction profiles from the content they produce in a social system. Such profiles are key to find the best person to contact based on an information ne...

Citations

... Content-based recommender systems require the end-user to describe using natural language or any other means like rating to show their preference [4][5][6][7][8]. This content obtained from the user's history uses to make a recommendation [9][10][11]. ...
... e core of the HGRVS algorithm is to first use the hypergraph sorting algorithm to classify English news text, and then select the items that need to be used from the already classified categories for backup, and compare the selected items above with other news relevance. And remove redundant parts to maintain complete coverage of news events [7]. Secondly, an English news text summarization algorithm based on hypergraph random walk is proposed. ...
... In such a context, customer opinion summarization and sentiment analysis (Ding, Liu, & Yu, 2008; Zhuang, Jing, & Zhu, 2006 ) techniques represent effective augmentations to traditional recommendation strategy, for example by not recommending items that receive a lot of negative feedbacks (Zhou et al., 2012). Indeed, a lot of attention is nowadays being payed from vendors to consumer's voices because of the great influence they may have on the opinions and decisions of others (Kurilovas, Juskeviciene, Kubilinskiene, & Serikoviene, 2014; Stan, Muhlenbach, & Largeron, 2014) and some companies already provide several opinion mining services (e.g., Amazon, Epinions, etc.). In recent times, some works have been proposed to extend traditional collaborative filtering with the use of sentiment analysis techniques, thus providing effective improvement to system performances (Drigas, Ioannidou, Kokkalia, & Lytras, 2014; Leung, Chan, & Chung, 2006): most of them make use of Part Of Speech (POS) tagging techniques and aim at refining standard collaborative filtering ranking outcomes in terms of numerical scales to take into account user community opinions. ...
... Some of these systems operate on DBpedia. DBpedia constitutes a valuable knowledge to solve the natural language polysemy by using contextual elements [172] e.g. recognizing that a web pages deals with the Apple company and not with the fruit. ...