Classification of Group Potency Levels of Software Development Student Teams

Castro-Hernandez, Alberto; Swigger, Kathleen; Serce, Fatma Cemile; López, Víctor

doi:10.17562/pb-51-8

Cited by 5 publications

(4 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This was achieved by focusing on specific factors that tend to affect team cohesion. These factors were derived from the research literature [19,20,46], and from smaller experiments performed by the author that looked at issues related to predicting the group constructs of Task cohesion [10,11].…”

Section: Methodsmentioning

confidence: 99%

“…These interactions can be analyzed and sorted by quantity, content and time; and can provide a rich data source for describing how a team's interaction can affect group work. Previous researchers have used this type of data to predict or describe a number of different group constructs such as leadership, cohesiveness, and group potency [3,42,10]. Moreover, some of the data has been used to describe behaviors that were related to effective team performance [45].…”

Section: Motivationmentioning

confidence: 99%

See 1 more Smart Citation

Content and Temporal Analysis of Communications to Predict Task Cohesion in Software Development Global Teams

Castro-Hernandez

2016

2016 IEEE 11th International Conference on Global Software Engineering Workshops (ICGSEW)

View full text Add to dashboard Cite

numbered references.Messages were collected from three software development projects that involved students from two different countries. The similarities and quantities of these interactions were computed and analyzed at individual and group levels. Results of interaction-based metrics showed that the collaboration variables most related to Task Cohesion were Linguistic Style Matching and Information Exchange. The study also found that Information Exchange rate and Reply rate have a significant and positive correlation to Task Cohesion, a factor used to describe participants' engagement in the global software development process. All these results suggest that metrics based on rate can be very useful for predicting cohesion in virtual groups. Similarly, content features based on communication categories were used to improve the identification of Task Cohesion levels. Also, at a group-level, all models were found correlated to Task Cohesion, specifically, Similarity+Rate, which suggests that models that include social and work communication categories are also good predictors of team cohesiveness. Finally, temporal interaction similarity measures were calculated to assess their prediction capabilities in a global setting. Results showed a significant negative correlation between the Pacing Rate and Task Cohesion, which suggests that frequent communications increases the cohesion between team Last but not least, I want to thank my family for always encouraging me and believing in me. This dissertation is also a result of their unconditional love and support.iii Another important team process that is affected by conditions within a remote collaboration setting is group cohesion. Previous research has found that communication technology (such as chat and email) often delays the development of the group cohesion construct [38]. As a result, global teams tend to have much lower group cohesion levels than co-located groups. One of the reasons why it is important to examine group cohesion levels within global virtual teams is that research has shown that this construct seems to affect how a team deals with different obstacles during the development of a project. In addition, the relation between group cohesion and other constructs (e.g. trust) has been shown to have an indirect relationship to team performance [21]. Using the above studies as a starting point, we studied the integration of metrics based on interaction type and amount, content features and temporal data to predict group cohesiveness.Our investigation used data collected in a collaborative setting in which participants from different countries worked together to complete a software development project.This study focused on the following aspects of the construct: first, it compared previous collaboration measures that had been used to analyze computer-based communications and determined which of these measures were most effective for predicting Task Cohesion; second, we developed a measure based on content features and determined if it could be used to im...

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Motivationmentioning

confidence: 99%

Content and Temporal Analysis of Communications to Predict Task Cohesion in Software Development Global Teams

Castro-Hernandez

2016

2016 IEEE 11th International Conference on Global Software Engineering Workshops (ICGSEW)

View full text Add to dashboard Cite

show abstract

“…Lee et al [32] developed a computational model of interpersonal trust through hand-coded nonverbal social behaviors and explored the temporal dynamic of the construct by means of hidden Markov models. Coming to team potency, Castro-Hernandez and colleagues [7] addressed it as both a regression and a classification problem, but their work was narrowly focused on virtual teams and required the timeconsuming recording of a dataset of students' interactions.…”

Section: Related Workmentioning

confidence: 99%

“…To annotate potency, we adopt the 8-items scale developed by Guzzo et al [24]. Specifically, in this study, the 7-point version of the scale (from To no extent (1) to To great extent (7)) is used [14].…”

Section: Annotationmentioning

confidence: 99%

Few Labels are Enough! Semi-supervised Graph Learning for Social Interaction

Corbellini,

Giraldo,

Varni

et al. 2023

2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)

View full text Add to dashboard Cite

Endowing machines with social intelligence is a fundamental goal of artificial social intelligence. Dealing with human-centered phenomena requires, however, a considerable amount of manually annotated data, making data annotation a costly and challenging task that hinders the training of supervised learning algorithms. In this study, we apply an approach grounded on Graph Convolutional Network (GCN) to alleviate the annotation burden. As a test bed, we select emergent states analysis with specific reference to the team potency. At first, we build the PO-TENCY dataset by fusing three datasets on social interaction. Next, we compute a set of multimodal features characterizing the social behavior of the team members and the team as one. Finally, we feed the POTENCY dataset to a semi-supervised GCN, trained on a binary node classification task, with variable amounts of labels. We show that GCN can assign team potency labels to an unlabeled team in the dataset by using only a few labeled examples (i.e., 10% of data), with performances comparable to or higher than those of two baseline algorithms carrying out the same task in a fully supervised way.

show abstract