Date of Graduation

5-2021

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Engineering (PhD)

Degree Level

Graduate

Department

Computer Science & Computer Engineering

Advisor

Xiaoqing "Frank" Liu and Brajendra Nath Panda

Committee Member

Douglas Adams

Second Committee Member

Susan E. Gauch

Third Committee Member

Justin Zhan

Fourth Committee Member

Qinghua Li

Keywords

Group Behavior, Group Interaction, Mashup Software, Opinion, Prediction, Recommendation

Abstract

Mashup application development is becoming a widespread software development practice due to its appeal for a shorter application development period. Application developers usually use web APIs from different sources to create a new streamlined service and provide various features to end-users. This kind of practice saves time, ensures reliability, accuracy, and security in the developed applications. Mashup application developers integrate these available APIs into their applications. Still, they have to go through thousands of available web APIs and chose only a few appropriate ones for their application. Recommending relevant web APIs might help application developers in this situation. However, very low API invocation from mashup applications creates a sparse mashup-web API dataset for the recommendation models to learn about the mashups and their web API invocation pattern. One research aims to analyze these mashup-specific critical issues, look for supplemental information in the mashup domain, and develop web API recommendation models for mashup applications. The developed recommendation model generates useful and accurate web APIs to reduce the impact of low API invocations in mashup application development.

Cyber-Argumentation platform also faces a similarly challenging issue. In large-scale cyber argumentation platforms, participants express their opinions, engage with one another, and respond to feedback and criticism from others in discussing important issues online. Argumentation analysis tools capture the collective intelligence of the participants and reveal hidden insights from the underlying discussions. However, such analysis requires that the issues have been thoroughly discussed and participant’s opinions are clearly expressed and understood. Participants typically focus only on a few ideas and leave others unacknowledged and underdiscussed. This generates a limited dataset to work with, resulting in an incomplete analysis of issues in the discussion. One solution to this problem would be to develop an opinion prediction model for cyber-argumentation. This model would predict participant’s opinions on different ideas that they have not explicitly engaged.

In cyber-argumentation, individuals interact with each other without any group coordination. However, the implicit group interaction can impact the participating user's opinion, attitude, and discussion outcome. One of the objectives of this research work is to analyze different group analytics in the cyber-argumentation environment. The objective is to design an experiment to inspect whether the critical concepts of the Social Identity Model of Deindividuation Effects (SIDE) are valid in our argumentation platform. This experiment can help us understand whether anonymity and group sense impact user's behavior in our platform. Another section is about developing group interaction models to help us understand different aspects of group interactions in the cyber-argumentation platform.

These research works can help develop web API recommendation models tailored for mashup-specific domains and opinion prediction models for the cyber-argumentation specific area. Primarily these models utilize domain-specific knowledge and integrate them with traditional prediction and recommendation approaches. Our work on group analytic can be seen as the initial steps to understand these group interactions.

Share

COinS