Nicole Novielli

The challenges of sentiment detection in the social programmer ecosystem

A recent research trend has emerged to study the role of affect in in the social programmer ecosystem, by applying sentiment analysis to the content available in sites such as GitHub and Stack Overflow. In this paper, we aim at assessing the suitability of a state-of-the-art sentiment analysis tool, already applied in social computing, for detecting affective expressions in Stack Overflow. We also aim at verifying the construct validity of choosing sentiment polarity and strength as an appropriate way to operationalize affective states in empirical studies on Stack Overflow. Finally, we underline the need to overcome the limitations induced by domain-dependent use of lexicon that may produce unreliable results.

Mining successful answers in stack overflow

Recent research has shown that drivers of success in online question answering encompass presentation quality as well as temporal and social aspects. Yet, we argue that also the emotional style of a technical contribution influences its perceived quality. In this paper, we investigate how Stack Overflow users can increase the chance of getting their answer accepted. We focus on actionable factors that can be acted upon by users when writing an answer and making comments. We found evidence that factors related to information presentation, time and affect all have an impact on the success of answers.

Anger and its direction in collaborative software development

Recent research has provided evidence that software developers experience a wide range of emotions. We argue that among those emotions anger deserves special attention as it can serve as an onset for tools supporting collaborative softwaredevelopment. This, however, requires a fine-grained model of the anger emotion, able to distinguish between anger directed towards self, others, and objects. Detecting anger towards self could be useful to support developers experiencing difficulties, detection of anger towards others might be helpful for community management, detecting anger towards objects might be helpful to recommend and prioritize improvements. As a first step towards automatic identification of anger direction, we built a classifier for anger direction, based on a manually annotated gold standard of 723 sentences that were obtained by mining comments in Apache issue reports.

The role of social media in affective trust building in customer---supplier relationships

Trust represents a key issue in building successful customer–supplier relationships. In this sense, social software represents a powerful means for fostering trust by establishing a direct, more personal communication channel with customers. Therefore, companies are now investing in social media for building their social digital brand and strengthening relationships with their customers. In this paper, we presented two experiments by means of which we investigated the role of traditional websites and social media in trust building along the cognitive and affective dimensions. We hypothesize that traditional websites (content-oriented) and social media (interaction-oriented) may have a different effect on trust building in customer–supplier relationships, based on the first impression provided to potential customers. Although additional research is still needed, our findings add to the existing body of evidence that both cognitive and affective trust can be successfully fostered through online presence. Specifically, social media provide companies with tools to communicate benevolence to potential customer and, therefore, foster the affective commitment of customers. Traditional websites, instead, are more appropriate for communicating the competence and reliability of a company, by fostering trust building along the cognitive dimension. The results of our studies provide implications for researchers and practitioners, by highlighting the importance of combining the two media for effectively building a trustworthy online company image.

Generating comparative descriptions of places of interest in the tourism domain

When visiting cities as tourists, most of the times people do not make very detailed plans and, when choosing where to go and what to seem they tend to select the area with the major number of interesting facilities. Therefore, it would be useful to support the user choice with contextual information presentation, information clustering and comparative explanations of places of potential interest in a given area. In this paper we illustrate how MyMap, a mobile recommender system in the Tourism domain, generates comparative descriptions to support users in making decisions about what to see, among relevant objects of interest.

Social Attitude Towards A Conversational Character

This paper describes our experience with the design, implementation and validation of a user model for adapting health promotion dialogs with ECAs to the attitude of users toward the agent. The model was conceived in agreement with the theory of social emotions in communication. It integrates a linguistic parser with a dynamic Bayesian network and was learnt from a corpus of data collected with a Wizard of Oz study

Sentiment Polarity Detection for Software Development

The role of sentiment analysis is increasingly emerging to study software developers’ emotions by mining crowd-generated content within social software engineering tools. However, off-the-shelf sentiment analysis tools have been trained on non-technical domains and general-purpose social media, thus resulting in misclassifications of technical jargon and problem reports. Here, we present Senti4SD, a classifier specifically trained to support sentiment analysis in developers’ communication channels. Senti4SD is trained and validated using a gold standard of Stack Overflow questions, answers, and comments manually annotated for sentiment polarity. It exploits a suite of both lexicon- and keyword-based features, as well as semantic features based on word embedding. With respect to a mainstream off-the-shelf tool, which we use as a baseline, Senti4SD reduces the misclassifications of neutral and positive posts as emotionally negative. To encourage replications, we release a lab package including the classifier, the word embedding space, and the gold standard with annotation guidelines.

Social robots and ECAs for accessing smart environments services

As far as interaction is concerned Ambient Intelligence (AmI) research emphasizes the need of natural and friendly interfaces for accessing services provided by the environment. In this paper we present the result of an experimental study aiming at understanding whether Embodied Conversational Agents (ECAs) and Social Robots may improve the naturalness and effectiveness of interaction by playing different roles when acting as interface between users and smart environment services. Results obtained so far show that ECAs seem to have a better evaluation than robots for information related tasks. On the other side, Social Robots are preferred for welcoming people and for guiding them in the smart environment, due to their possibility to move and to the perceived sense of presence. Moreover, the robot seems to elicit a more positive evaluation in terms of user experience.

How to ask for technical help? Evidence-based guidelines for writing questions on Stack Overflow

Abstract Context The success of Stack Overflow and other community-based question-and-answer (Q&A) sites depends mainly on the will of their members to answer others’ questions. In fact, when formulating requests on Q&A sites, we are not simply seeking for information. Instead, we are also asking for other peoples help and feedback. Understanding the dynamics of the participation in Q&A communities is essential to improve the value of crowdsourced knowledge. Objective In this paper, we investigate how information seekers can increase the chance of eliciting a successful answer to their questions on Stack Overflow by focusing on the following actionable factors: affect, presentation quality, and time. Method We develop a conceptual framework of factors potentially influencing the success of questions in Stack Overflow. We quantitatively analyze a set of over 87 K questions from the official Stack Overflow dump to assess the impact of actionable factors on the success of technical requests. The information seeker reputation is included as a control factor. Furthermore, to understand the role played by affective states in the success of questions, we qualitatively analyze questions containing positive and negative emotions. Finally, a survey is conducted to understand how Stack Overflow users perceive the guideline suggestions for writing questions. Results We found that regardless of user reputation, successful questions are short, contain code snippets, and do not abuse with uppercase characters. As regards affect, successful questions adopt a neutral emotional style. Conclusion We provide evidence-based guidelines for writing effective questions on Stack Overflow that software engineers can follow to increase the chance of getting technical help. As for the role of affect, we empirically confirmed community guidelines that suggest avoiding rudeness in question writing.

UNIBA: Sentiment Analysis of English Tweets Combining Micro-blogging, Lexicon and Semantic Features

This paper describes the UNIBA team participation in the Sentiment Analysis in Twitter task (Task 10) at SemEval-2015. We propose a supervised approach relying on keyword, lexicon and micro-blogging features as well as representation of tweets in a word space.