Citation Context

...entiStrength web site to try out SentiStrength live online. The web site includes the main English version as well as several other language variants. SentiStrength’s commercial users include Yahoo! (=-=Kucuktunc, Cambazoglu, Weber, & Ferhatosmanoglu, 2012-=-; Weber, Ukkonen, & Gionis, 2012) and a range of online information management companies around the world. It was also used to power a light display on the EDF Energy London Eye during the London 2012...

Citation Context

...che httpd server mailing list [34]. They analyzed the personality of four top developers, assessing the personality of two top developers who have left the project. Bazelli et al. replicated Rigby’s work on Stack Overflow to analyze the personality traits of different authors [4]. Although both studies deal with the behaviour of developers, they do not consider the emotions of project members and how these impact the motivation of project members. Typically, emotion and sentiment mining studies focus on emotions that people convey through twitter, question and answer sites and product reviews [28, 24, 25]. Choudhury et al. [9] applied sentiment analysis on microblogging data of a large global Fortune 500 software corporation used to “disseminate product/service updates to the larger enterprise community, such as details of new upcoming features, news about future team meetings, releases or trade shows”. They found that there is a potential for building systems that assess feelings of employees in an enterprise. To improve emotional awareness in software development teams, Guzman et al. proposed a sentiment analysis approach for discussions in mailing lists and web-based software collaboration ...

Citation Context

... text of each comment has been processed using SentiStrength [28], a sentiment analysis tool for the extraction of emotional content from short texts. SentiStrength is the state-of-the-art tool for lexiconbased analysis of short, informal text. This is the case of Youtube comments, for which SentiStrength’s accuracy is above 88% when classifying positive from negative text [27]. In general, it achieves human accuracy, i.e. it can estimate emotional content from text with a similar error rate as a human reader. It has been recently used for the analysis of emotional expression in Yahoo answers [14], Twitter trends [26], and chatroom communication [9]. SentiStrength uses a lexicon of emotional-bearing terms combined with the detection of negations, amplifiers and diminishers, providing two values, a positive score p ∈ [+1,+5], and a negative score n ∈ [−1,−5]. As the text of Youtube comments is very short (max. 500 characters), we aggregate both values to a measure of polarity. A comment c is classified as positive (ec = +1) if p+ n > 0, negative (ec = −1) if p+ n < 0, or neutral (ec = 0) if p = n and both have an absolute value lower than 4. This way, comments with high and equal positi...

Citation Context

...nt, by considering a small window of text around the mention (we include the ten words preceding the mention, and the ten words following it). We further calculate attitude and sentimentality metrics =-=[25]-=-, which measure “the inclination towards positive or negative sentiments” and the “amount of sentiment,” respectively. Finally, for every entity we compute average attitude and sentimentality over all...

Citation Context

...line communities [7]. Among its previous applications, Sentistrength has been used to analyze emotions about political topics in Youtube and Twitter [28, 29], product reviews [23], and Yahoo! answers =-=[30]-=-. 2.3 Social Resilience and Contributor Motivation The question of how groups are formed and disappear has been addressed for online social networks and scientific communities [31, 32]. This highlight...

Citation Context

...focused on identifying the linguistic features that contribute to the affective content of text and how these features could be automatically extracted to obtain a sentiment metric. In particular, in =-=[24]-=-, two complementary sentiment metrics were defined: sentimentality, which quantifies the total amount of sentiments expressed in the content, and polarity (attitude), which represents the inclination ...

Citation Context

...e argued and proved that among the nontechnical factors, which can influence the members of online communities, the emotional style of a technical question does affect the probability of promptly obtaining a satisfying answer [29], [9]. More specifically, our effort is to understand how expressing affective states in Stack Overflow influences the probability for askers of eliciting an accepted answer and the probability for answerers of having an answer accepted. Our research follows a recent trend that has emerged to study the role of affect in social computing. For example, Kucuktunc et al. [19] performed a large-scale sentiment analysis study on Yahoo! Answers to assess the impact of the semantic orientation of a post on its perceived quality. Althoff et al. [1] found that expressing gratitude in a question is positively correlated with success of altruistic requests in Reddit.com. Guzman et al. [17] perform sentiment analysis of commit comments in GitHub and demonstrate that a correlation exists between emotions and other factors such as the programming language used in a project, the geographical distribution of the team and the day of the week. Similarly, Guzman and Bruegge [16] ...

Citation Context

...ty as a proxy for similarity, and is the first paper to contribute geographic user partitioning for collaborative filtering. Our contribution contra LARS is that geographic user partitions need not be contiguous nor a compact space. 2.2.2 Location as “just another feature” Still other work in general recommender research treats location information as “just another feature”. Seroussi et al. [20] use standard matrix factorization techniques to find latent relevant features in a person’s preferences, with U.S. state names drawn from postal codes as possible features. Similarly, Kucuktunc et al. [12] match postal codes with census data, inferring demographics to match users in a question-and-answer site. We see promise in Kucuktunc’s method, which uses postal codes as a way to guess demographic tags for users. Their method is based on Weber and Castillo’s demographic analysis which is meant to “depersonalize and de-localize” location information [24], yet demographics are themselves geographically clustered [4], resulting in a “distant neighbors” type of approach through a demographic step. This approach has the benefit of accounting for locations a recommender has never before encountere...