A seasoned, qualified and accomplished IT professional with over 13 years of quality work experience in several challenging roles ranging from software analysis, design, development and managing software projects in various countries including Kenya, Ethiopia, Malawi, Tunisia, Libya, Morocco, Algeria, South Sudan and Nepal. Stephen is a highly committed and self-driven professional who is able to work in diverse environments, adapt to different cultures and achieve outstanding work results. Stephen has worked with clients from different industries including retail, manufacturing, humanitarian organizations, non-governmental organizations and government institutions. Stephen has experience in working with GIS based applications geared towards solving problems that are directly related or caused by environmental factors and also has developed tools that aid the humanitarian organizations track the effectiveness and impacts of social projects and social spending. Some of the software development technologies include Python, C++, ASP.Net, C#, Javascript, ReactJS, CSS, JQuery, ExtJs, MySQL, Postgres. Stephen holds a Bachelor’s degree in Computer Science from the University of Nairobi and is a qualified accountant holding CPA section VI.

Project Summary

Project  Title: A Model for Processing Public Participation Feedback Using Topic Modeling (A Case for Public Task Forces in Kenya)

Research Supervisor: Dr. Lawrence Muchemi

Abstract: Governments acknowledge the need to involve citizens, through different platforms like public task forces, when public policies are drafted. This paper focuses on public task forces as one of the platforms for citizen engagement. The amount of feedback received is beyond the task forces’ processing ability which causes input from critical stakeholders being ignored completely leading to biased output. In this paper, we propose a model that task forces can use to process feedback received from stakeholders. Specifically, the model identifies topics contained in submissions by applying topic modeling using LDA algorithm. 22 submissions to an ICT procurement task force were used in developing the model. To validate the model, topics identified by the model were compared against those identified by a human expert. Results show that the model generated topics that are similar to topics identified by a human expert. This paper contributes to practice by enabling task forces to objectively identify topics covered by submissions which results into better acceptance of outputs and recommendations of task forces by the citizens. The model can be generalized to other establishments like constitutional commissions. The results of this study show that topic modeling can be applied in influencing the outcomes and quality of recommendations of task forces thus improving governance by ensuring that citizens’ views are adequately factored.