Automating Forms creation using AI
Ruwodo, David (2024-07-29)
Automating Forms creation using AI
(29.07.2024)
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
avoin
Julkaisun pysyvä osoite on:
https://urn.fi/URN:NBN:fi-fe2024080263388
https://urn.fi/URN:NBN:fi-fe2024080263388
Tiivistelmä
This study investigates the potential of artificial intelligence (AI) in automating form creation within software development, focusing on the interpretation of functional specification documents for automated form generation. The research addresses the inefficiencies in current form creation practices, particularly for organizations with dynamic requirements and contract programmers facing resource constraints.
The study employs a multifaceted methodology, including a comprehensive literature review on the historical progression of AI in software development and a technological feasibility study. The research explores the capabilities of Large Language Models (LLMs) in formatting raw functional specifications and generating synthetic form components.
Key findings reveal that while LLMs show promise in handling simple form components and small-scale generation tasks, they struggle with complex relationships and full-form generation. The study evaluates various fine-tuning techniques and their effectiveness across different models, highlighting the importance of high-quality, task-specific training data.
Results indicate that while AI demonstrates potential in certain aspects of form creation, current models fall short of producing production-ready code for complex forms. The research also uncovers unexpected performance variations between model sizes and the effectiveness of language-specific models.
This study contributes to the growing body of knowledge on AI-assisted software development, offering insights into the current capabilities and limitations of AI in form creation. It concludes by suggesting future research directions and practical implications for both researchers and practitioners in the field.
The study employs a multifaceted methodology, including a comprehensive literature review on the historical progression of AI in software development and a technological feasibility study. The research explores the capabilities of Large Language Models (LLMs) in formatting raw functional specifications and generating synthetic form components.
Key findings reveal that while LLMs show promise in handling simple form components and small-scale generation tasks, they struggle with complex relationships and full-form generation. The study evaluates various fine-tuning techniques and their effectiveness across different models, highlighting the importance of high-quality, task-specific training data.
Results indicate that while AI demonstrates potential in certain aspects of form creation, current models fall short of producing production-ready code for complex forms. The research also uncovers unexpected performance variations between model sizes and the effectiveness of language-specific models.
This study contributes to the growing body of knowledge on AI-assisted software development, offering insights into the current capabilities and limitations of AI in form creation. It concludes by suggesting future research directions and practical implications for both researchers and practitioners in the field.