Project Overview
Project Summary:
Our project has created an Android app that harnesses employment data to provide personalized job recommendations for individuals with disabilities. Leveraging Auto machine learning, our system considers age, location, disability type, and severity to identify suitable employment opportunities. The app is enhanced with STT and TTS capabilities to support users with different communication needs and includes a specialized community platform for shared experiences. By integrating job description videos through a Text-To-Video API, the app deepens users' role understanding and engagement. This service is designed to optimize job matching and increase workplace accessibility, marking a significant advancement in the employment sector for the disabled community. While developing Android app, we designed a detailed business model and conducted a SWOT analysis to prepare for actual service.
Identifying the Challenge
The Social Problem:
The proportion of disabled individuals within the total population is on the rise, yet in South Korea, not a single application dedicated to their needs exists. Despite continuous efforts to implement disability employment projects, employment challenges for the disabled persist. Analysis of disability employment-related complaints from 2017 to 2019 reveals not only a persistent shortage of job placements but also a lack of suitable employment opportunities. Regular employment sites offer limited options for the disabled, highlighting a clear need for targeted job facilitation services.
Innovation and Uniqueness
Why Our Project Stands Out:
Our project stands out by integrating AutoML to design a high-accuracy job recommendation system tailored to disabled workers' data. We've further innovated by preprocessing job description data with GPT and generating detailed job videos using Text-To-Video-api model. This approach provides intuitive and accessible job description videos to disabled users, significantly enhancing the job-search experience over traditional text-based systems. And we made the project more realistic by designing a feasible business model.
Insights and Development
Learning Journey:
In our project, students from distinct academic disciplines joined forces, initially grappling with communication barriers due to the varied expertise. Embracing our individual strengths and cultivating a collaborative ethos, we surmounted these hurdles, ultimately forging a project of high caliber. This interdisciplinary endeavor did not just enhance the project outcome; it unexpectedly broadened our collective knowledge, with each member gaining insights from the others' specialized academic domains.
Development Process:
Our project was meticulously crafted using Android Studio for app development, with AutoML and machine learning forming the backbone of our job recommendation engine. The process began with a thorough analysis of disability employment data, leading to the design phase where the app's concept was shaped to address the identified employment disparities. We developed a prototype that utilized Text-To-Video technology to produce comprehensible job description videos. Through iterative testing and refinement, we honed our prototype into a fully functional app. Collaboration was key; our multidisciplinary team employed agile methodologies to rapidly prototype, test, and iterate, ensuring a problem-solving approach that was both dynamic and user centric.
