Content area
Nowadays, effective document management is crucial for enhancing organizational productivity, efficiency, and adaptability, but traditional document management systems often struggle with retrieval, accessibility, and workflow automation. Thus, the present research introduces an Automated Document Management System (ADMS) prototype that integrates ChatGPT as an intelligent assistant to enhance document processing. By using natural language processing (NLP) and machine learning, the system enables users to interact with documents through conversational queries, analysis, retrieval, and summarisation requests. The ChatGPT-powered assistant operates by creating small context sessions to dynamically analyse document content. These sessions extract key insights, highlight relevant sections, and generate summaries according to user queries. By maintaining context within each session, the assistant delivers accurate and relevant responses while ensuring continuity in the conversation. This approach not only improves document accessibility and usability but also enables the integration of other knowledge domains within an organization. To evaluate its effectiveness, the prototype is validated by 30 IT experts, who identify essential functionalities, strengths, and potential challenges. Their feedback provides key insights into areas such as accuracy, scalability, and security, guiding future improvements. This research aims to demonstrate how ChatGPT can optimise information handling and support decision-making, highlighting the revolutionary potential of AI-driven document management. By automating process and project definition and enhancing search capabilities, the system offers a scalable and intelligent solution for modern document workflows. Thus, Knowledge Management (KM) evolves to focus on contextual understanding, emphasises cohesive data integration, and expands the scope of knowledge creation and sharing and their influence on knowledge domains.
Abstract: Nowadays, effective document management is crucial for enhancing organizational productivity, efficiency, and adaptability, but traditional document management systems often struggle with retrieval, accessibility, and workflow automation. Thus, the present research introduces an Automated Document Management System (ADMS) prototype that integrates ChatGPT as an intelligent assistant to enhance document processing. By using natural language processing (NLP) and machine learning, the system enables users to interact with documents through conversational queries, analysis, retrieval, and summarisation requests. The ChatGPT-powered assistant operates by creating small context sessions to dynamically analyse document content. These sessions extract key insights, highlight relevant sections, and generate summaries according to user queries. By maintaining context within each session, the assistant delivers accurate and relevant responses while ensuring continuity in the conversation. This approach not only improves document accessibility and usability but also enables the integration of other knowledge domains within an organization. To evaluate its effectiveness, the prototype is validated by 30 IT experts, who identify essential functionalities, strengths, and potential challenges. Their feedback provides key insights into areas such as accuracy, scalability, and security, guiding future improvements. This research aims to demonstrate how ChatGPT can optimise information handling and support decision-making, highlighting the revolutionary potential of AI-driven document management. By automating process and project definition and enhancing search capabilities, the system offers a scalable and intelligent solution for modern document workflows. Thus, Knowledge Management (KM) evolves to focus on contextual understanding, emphasises cohesive data integration, and expands the scope of knowledge creation and sharing and their influence on knowledge domains.
Keywords: Knowledge management, Document management, Prototype, Research, ChatGPT, AI
1. Introduction
Effective documentation is essential for knowledge management and knowledge sharing in organizations, as it ensures the secure storage and access to critical information and acts as a strategic tool in decision-making. Employees are an essential asset to any organization, accumulating knowledge and skills that must be documented to avoid loss upon their departure (Navidi et al., 2019). Well-structured documentation management (DM) supports the transformation of tacit knowledge into explicit knowledge, facilitating the sharing of experience and best practices within the organization. However, various studies highlight challenges related to knowledge documentation, including the lack of detail about methodologies and errors, which limits the ability to learn from experience, and the need for having a Documentation Management System (DMS) as a technological approach to categorizing and providing analysis of different organizational assets (Nilekar, 2024).
The present research aims to investigate the specifics of KM in software processes and to propose an Automated Documentation Management System that integrates tools based on artificial intelligence (AI). The integration of AI into corporate documentation processes has the potential to substantially improve the efficiency of knowledge management. This advancement aligns with and can be further reinforced by the established paradigm of conversational agents, such as chatbots, which facilitate interactive access to document content and organizational knowledge (Tin et al., 2024). By documenting business processes and projects, a clear model for understanding and transferring knowledge can be created, facilitating the training of new employees and optimizing company workflows (Haddadpoor et al., 2015). AI tools can facilitate multiple DM processes, providing automation of analysis, categorization, and linking information, providing access, and improving the knowledge flow within the organization. AI-based systems can support technology integration through intelligent search engines, automated summaries, and process analysis, thereby facilitating business process reengineering. AI-enriched documentation ensures quick access to relevant information and improves an organization's adaptability to new requirements and challenges.
The paper is structured as follows. The background part describes the main theoretical foundations, serving as a starting point in the system design. Next, the prototype of the ADMS system is presented, exploring first the system requirements and then outlining the layers of the prototype design. Implementation and validation sections provide relevant insights about the users' perceptions of the ADMS system. Finally, the conclusion outlines the perspectives and future directions for implementing AI tools in document management systems.
2. Background
Document management systems play a crucial role in optimizing organizational processes and promoting collaboration. Documentation management is often perceived as a challenging task due to the inaccessibility, unstructured nature, and complexity of the company information. This results in difficulties integrating new employees, reduced workflow efficiency, and slower innovation adoption. By implementing standardized processes and procedures, company documentation helps ensure compliance with regulatory requirements and industry standards, reducing legal risks and preventing errors. It provides clarity in procedures, ensuring that employees follow established rules. In regulated industries, such as healthcare and finance, documentation demonstrates compliance with legislation and facilitates audits (von der Heyde & Breiter, 2017). In this sense, documentation serves as the backbone of KM in an organization, supporting its ability to absorb new ideas, adapt to change, and improve internal processes. AI can enhance and optimize DM processes. According to several studies, sentiment analysis techniques (such as text classification and recurrent neural networks (RNN)) significantly improve the processing of textual information (Lim et al., 2020). Features such as document search, messaging, and version control in DMS stand out as key factors for more effective collaboration and workflow improvement (Regla & Marquez, 2020), leading to increased system adoption (Ayaz and Yanartaş, 2020). For example, the design and implementation of an electronic document management system (EDMS), tailored to the specific needs of educational institutions, has a transformative impact on administrative processes and collaboration (Mahmood, 2017).
To respond to these challenges with opportunities provided by AI, the authors propose an Automated Documentation Management System (ADMS) design, aiming to provide a centralized and intelligent approach for managing technical and project documentation. By leveraging AI and using a systematic model of KM in Software Engineering (SE), the system focuses on incorporating functions in the following knowledge domains (Georgiev, 2023):
* Technocratic knowledge - technology portfolio, experience in product development, information systems, and innovations;
* Behavioural knowledge - organizational culture, internal policies, and practices;
* Product knowledge - functionality and features of the developed products;
* Process knowledge - documentation on methodologies, workflows, and best practices.
Various researchers emphasize the essential role of documentation in maintaining human oversight over AI systems and supporting the evaluation of human decision-making impact on business processes. Creating comprehensible documentation through AI ensures compliance with regulatory standards (Königstorfer & Thalmann, 2022). For instance, enhancing the requirements elicitation process can be achieved by bridging the gap between AI strategies (including machine learning and natural language processing) and traditional methods that rely on human interaction and manual procedures (Cheligeer et al., 2022; Chomal et al., 2024). The adoption of AI tools as part of KM initiatives can significantly improve knowledge processes by integrating technological innovation and effective requirements engineering techniques (Cerchione & Esposito, 2017). Examples of AI-based solutions include:
* ChatGPT - provides intelligent search and personalized responses, helping developers quickly locate precise information instead of spending time sifting through bulky documents. AI can also analyse recurring queries and suggest improvements in knowledge management systems.
* GitHub Copilot - offers automatic code snippets that accelerate development and deployment processes while preventing errors.
* DeepCode - analyses the codebase to identify bugs, optimize performance, and detect potential security risks.
* Microsoft Power BI and Google Cloud AutoML - analyse large volumes of data, identifying trends and patterns that support better-informed decision-making and workflow optimization.
Beyond these tools, AI can also be used for automatic test case generation, improving employee onboarding processes, and offering personalized recommendations for training and development.
The implementation of AI solutions leads to more effective KM by not only automating routine tasks but also reducing the risk of losing critical information. In the long run, these technologies will play a central role in transforming workflows, ensuring better connectivity, accessibility, and adaptability of knowledge within an organization.
3. Prototype Design and Main Functionalities
3.1 ADMS Functionalities
The prototype of the ADMS aims to demonstrate the applicability and effectiveness of intelligent agents and AI automation to improve the accessibility, understanding, and management of complex software documentation. Thus, the ADMS prototype proposes structured, accessible, and automated mechanisms for storing, analysing, and utilizing documentation derived from the main knowledge domains (Georgiev & Antonova, 2024). It employs AI tools to recognise key concepts, extract new knowledge, and improve navigation across information repositories.
The core ADMS modules support the following key capabilities:
* "My Documents" module provides users with a personal space for storing, uploading, downloading, organizing, and analysing documents using an intelligent assistant, capable of generating summaries and classifying files.
* "Processes" module enables the definition of various workflows, including operational, procedural, and project-related processes, to support software development and organizational governance. It introduces the definition of multiple steps, each with an associated template file that users can download, complete, and submit as part of the process.
* "Projects" module allows users to initiate and manage projects by setting parameters such as budget and timeline, linking them to predefined processes for consistency, managing project documentation with a folder tree, and using an AI assistant for document summarization and content analysis.
Among the most important non-functional requirements for the ADMS are security, performance, an intuitive user interface, and the models of integration with intelligent assistants. The authorization modules ensure that only authorized users have access to sensitive documents, processes, and project information. The system supports common file formats (PDF, DOCX, TXT, XLSX, CSV) to allow text data extraction, analysis, and integration with the intelligent assistant. This is essential, as various business processes involve working with text documents, forms, and structured files.
The intuitive interface of the ADMS enables easy navigation, reducing onboarding time and increasing user efficiency. The integration of an intelligent assistant is ensured by implementing the following key conditions:
* The assistant must be integrated via an API, providing standardized communication between the prototype and the external AI model.
* The system must send and receive data in structured formats (JSON/XML), with support for validation and result processing.
When considering the need for training the assistant and storing the history of uploaded documents, the following aspects must be considered (Table 1):
* Security and Usage Policies: Security measures include data encryption, regulatory compliance, and access restrictions to sensitive data (Goel et al., 2024; Sebastian, 2023).
ChatGPT is an appropriate model for developing an ADMS prototype due to its powerful GPT-4 engine, flexible API integration, and strong data security policies, including AES-256 encryption and GDPR compliance. It supports fine-tuning and file uploads, enabling custom document analysis and adaptive learning from organizational data. This provides a solid foundation to develop further functionalities, considering the assistants' specialisation. Unlike more limited ecosystem-bound models like Microsoft Copilot, ChatGPT offers broad compatibility with platforms, ensuring seamless integration into diverse workflows. This versatility, combined with its high-quality natural language capabilities, makes ChatGPT especially suitable for intelligent document categorisation, summarization across project management environments.
3.2 Prototype Design
The ADMS prototype is designed as a scalable and secure web-based solution that facilitates the management of documents, processes, and projects within an organization. Its architecture follows a multi-layered approach, with the main components including the client side (Client), server side (API), and database.
The client side is a dynamic Single Page Application (SPA) developed with Angular. It provides a responsive and user-friendly interface that allows access to the system's main modules. Each user action is sent to the server side via RESTful APIs. Access to the various modules depends on the user's role, and sessions are managed using JSON Web Tokens (JWT) for enhanced security.
The server side is implemented as a .NET API component, which provides the business logic and data management. The system exposes RESTful API endpoints that allow operations related to user, role, process, and project management with their associated documents. This layer handles authentication and authorization mechanisms.
When an access request is made, the user is authenticated via a username and password. Upon successful validation, the system generates a token containing information about the user's role. Based on this role, access levels and permitted operations are determined.
The service layer defines how operations are performed and how data is validated within the system. It provides functionalities for creating, reading, editing, and deleting users, roles, processes, projects, and documents. This layer also handles operations related to managing tree structures, which are essential for organizing documents into personalized spaces and project repositories.
The domain Layer defines the abstractions that describe the various classes and objects corresponding to the tables in the database. It is accessible by all other layers of the system, providing a unified structure for working with data.
Microsoft SQL Server (MSSQL) is used for data management. The database is designed to support:
* Hierarchical structures for organizing documents and processes;
* Related tables for storing users, roles, access rights, and project parameters;
* Indexed structure for fast data search and retrieval.
4. Prototype Implementation and Validation
The developed ADMS prototype implements a limited but functionally significant scope of the capabilities of the DMS (Document Management System), focusing on the key knowledge areas essential to the domain of Software engineering. The main components of the user interface include:
* Home Page - serves as the entry point to the system, providing users with access to different modules. Navigation is structured around the core functional areas.
* User Management - allows for viewing existing users and creating new ones. This implies the presence of role-based access control (RBAC) and administrative authority over system access.
* User Document Management - offers the ability to create folders, upload, and organize documents. This functionality supports structured storage and quick access to content according to individual user needs.
* Process Management - includes viewing and creating processes, showcasing the system's ability to define and manage operational, organizational, and project workflows. This supports standardization and traceability of actions.
* Project Management - enables the creation and viewing of projects, with the ability to attach documents and link them to predefined processes. Such functionality demonstrates the DMS's orientation toward a project-based approach to knowledge and documentation management.
4.1 UI Components
4.1.1 Main views
The administrator interface includes an "Administration" module for managing users, roles, and access rights across the system (Figure 1). The administrator can view detailed system status information, manage a personal file workspace, and configure or edit processes, marking whether they are system-wide (accessible to other users). Additionally, administrators can oversee project structures, attach relevant documentation, and ensure consistency in workflow execution.
Regular users have access to similar core functionalities, including managing their own file workspaces and editing processes, with the option to indicate system-wide availability. They can also manage individual projects, upload or attach related files, and collaborate with team members throughout the project lifecycle.
4.1.2 Documentation management component
Documentation management is a critical functionality supporting both the "My Documents" and "Projects" modules, acting as a central hub for organizing, accessing, and leveraging content throughout the system. It merges the advantages of traditional document management - covering essential CRUDD operations (Create, Read, Update, Delete, Download) - with advanced AI-driven capabilities for document understanding and analysis, powered by OpenAI technologies.
On the one hand is a robust file explorer, which enables users to create folders, upload and download files, and seamlessly interact with an intelligent assistant via "Summarise" and "Analyse" commands. This assistant, integrated directly into the file tree, can be triggered to perform operations and answer user queries. Through a conversational interface, users can ask specific questions or issue commands, enabling dynamic exploration of document content and extracting meaningful insights in real time.
Furthermore, the module offers the flexibility to choose between different OpenAI models, allowing users to tailor the assistant's capabilities based on performance needs or context. This integration of AI within a userfriendly document interface significantly enhances productivity, supports knowledge discovery, and positions the documentation management module as a powerful, intelligent feature across both personal and collaborative project spaces (Figure 2).
4.1.3 Wizard Component for Defining Process/Project Steps
The wizard component is a key element used to define process steps and automatically generate corresponding project steps based on predefined workflows (Figure 3). It provides a user-friendly interface that allows intuitive configuration and customization of each step's view, ensuring consistency and reusability even after a project has been initiated. This modular approach supports dynamic adaptation to different project needs and encourages process standardization across teams.
Project information is organized into three main sections for better clarity and navigation:
* "Project Information", which contains key metadata and configurations.
* "Gantt Chart", offering a visual representation of the timeline and dependencies between steps.
* "Attached Documents", which consolidates all relevant files and resources associated with the project.
4.2 ADMS Validation
The study followed a three-stage research methodology to evaluate the prototype. First, a set of testing scenarios is designed to simulate typical user interactions with the system. In the second stage, the respondents execute these scenarios to assess the system's functionality and usability. Finally, participants complete a structured feedback document and a survey, providing both qualitative impressions and quantitative ratings of their experience.
Several of the survey questions are based on a 5-point Likert scale, enabling further statistical analysis. For selected items, one-sample Student's t-tests are conducted by setting up a null hypothesis (H0: µ = 3), which represents a neutral position, suggesting the statement is not statistically supported. An alternative hypothesis (H1: µ > 3) is defined to test whether respondents' ratings significantly leaned in favour of the statement, indicating positive user alignment.
A total of 30 specialists from diverse professional backgrounds within the software industry participate in the ADMS prototype validation experiment. They test the system by executing several scenarios for document management and completing a short survey. The largest share of respondents (70%) are software developers, which adds additional expertise in terms of efficiency, performance, and design. The inclusion of project managers (23%), team leaders (17%), DevOps (10%), and Quality Assurance (7%) specialists contributes to a multifaceted assessment of the applicability of the ADMS in the knowledge areas.
The respondents have different levels of professional experience, with the largest share being specialists with between 5 and 10 years. They make up 50% of the participants, which suggests that the feedback is based on in-depth practical knowledge and experience in document, process, and project management.
The largest percentage of participants work in companies with 100 to 500 employees, which is indicative of medium-sized organizations with well-established processes. Respondents from small teams with less than 20 people, as well as from large enterprises with over 500 employees, are also included, which provides diverse perspectives on the usability of the ADMS in different corporate structures.
During the testing phase, respondents rate the various functionalities of the ADMS in terms of their usefulness. The results show that the intelligent assistant for document analysis has the highest value for users (83%). This highlights the importance of automated document analysis and summarization as a key functionality of the system. Next comes the Intelligent Project Assistant, as 70% of the users see value in supporting document management within projects. Both Project documentation and Project creation are ranked by 57% of the respondents as especially useful, highlighting the importance of effective document organization in the context of project activities and being significant for workflow management. The possibility to upload forms and define processes is useful for 37%, but the results show that they tend to be more specific to certain user groups.
The results of the survey show a high level of acceptance and positive attitude towards the implementation of the system. The numerical data are analysed by applying t-tests, which reject the null hypothesis H0: µ = 3 and accept the alternative hypothesis H1: µ > 3 (Table 2).
Respondents' feedback identifies key functionalities that would contribute to improving the system and its effectiveness in a real-world environment:
* Automatic report generation (73%) - would save significant time and reduce human error by compiling data and insights instantly. It ensures consistency in documentation, supports decisionmaking, and improves transparency across projects.
* Integration with external systems (70%) - tools like version control, ticketing systems, or HR platforms would allow seamless data exchange. This would increase the overall productivity, avoid data duplication, and create a unified workflow across different organizational tools.
* Automatic document categorization (57%) - would help with organizing knowledge efficiently, improve navigation, and enable faster retrieval of relevant content. It would enhance structure and reduce the cognitive load on users managing large repositories.
* Access control and rights (63%) - role-based access ensures sensitive information is protected while maintaining operational flexibility. The administrator module would be expanded with collaboration functionalities - sending invitations or dedicating roles for a given project. It minimizes security risks, ensures compliance with data policies, and supports tailored user experiences.
* Advanced search capabilities (50%) - powered by metadata and AI, advanced search would allow users to quickly find precise information. It boosts efficiency, reduces downtime, and supports better knowledge reuse and onboarding.
* Integrated collaborative chat (47%) - would enhance real-time collaboration, enabling users to discuss documents directly within the platform. This fosters faster feedback loops, improves decisionmaking, and captures knowledge from informal exchanges.
These results demonstrate that the implementation of the system is perceived as a positive step towards process optimization, although some users may have reservations about its long-term impact. The feedback confirms that the prototype successfully performs its main functions - document analysis, process and project management, and integration with the intelligent assistant. Users appreciate the extracted information, the lack of hallucinations, and the intuitive UX/UI interface.
5. Conclusion
The integration of AI is becoming a mandatory element of any knowledge management system that supports the implementation of software processes. In the context of documentation management, ADMS could facilitate access to various types of documentation by integrating intelligent agents within its functionalities.
The presented prototype of the ADMS enables validation of the system's key functionalities, such as documentation management across different knowledge domains, intelligent search and content analysis, and user access control. It will also verify the integration with intelligent agents and the available configuration options. This prototype provides a solid foundation for the further development of the ADMS and implementation of AI into a real-world technological solution supporting effective KM in SE.
Considering that the ADMS prototype demonstrates the potential influence of AI on the efficiency of various professionals in the software industry by supporting documentation storage, application insights, and the sharing of information. Thus, the model could be developed and deployed as a strategic tool that facilitates КМ for improving knowledge storage and communication within the organisations.
Acknowledgements
The authors gratefully acknowledge the support provided by the project UNITe BG16RFPR002-1.014-0004 funded by PRIDST.
Ethics Declaration: This study involved a voluntary survey of 30 participants, collecting only non-identifiable opinion-based responses. No personal or sensitive data was gathered. Participants were informed about the purpose of the research, and their consent was implied through participation. As such, formal ethical clearance was not required under the relevant institutional guidelines.
AI Declaration: Artificial intelligence tools were used solely for editing and language polishing. No AI tools were used for data generation, analysis, or interpretation.
References
Ayaz A. & Yanartaş M. (2020). An analysis on the unified theory of acceptance and use of technology theory (UTAUT): Acceptance of electronic document management system (EDMS). Computers in Human Behaviour Reports, Volume 2, 2020, 100032, ISSN 2451-9588, https://doi.org/10.1016/j.chbr.2020.100032
Cerchione R. & Esposito E. (2017). Using knowledge management systems: A taxonomy of SME strategies, International Journal of Information Management, Volume 37, Issue 1, Part B, 2017, Pages 1551-1562, ISSN 0268-4012, https://doi.org/10.1016/j.ijinfomgt.2016.10.007
Cheligeer C., Huang J., Wu G., Bhuiyan N., Xu Y., Zeng Y. (2022). Machine learning in requirements elicitation: a literature review. Artificial Intelligence for Engineering Design, Analysis and Manufacturing. 2022; 36:e32. doi:10.1017/S0890060422000166
Chomal, V., Patel J., Shah I., Solanki, B. (2024). AI-Driven Software Requirements Elicitation: A Novel Approach, Madhya Pradesh Journal of Social Sciences (A Biannual Journal of M.P Institute of Social Science Research, Ujjain) ISSN: 0973- 855X, Vol. 28, No.2, 2023
Georgiev, D. (2023). Exploring knowledge management from a software engineering perspective, European Conference on Knowledge Management, 2023, Volume 2: pp 1571-1578, 10.34190/eckm.24.2.1497
Georgiev, D., and Antonova, A. (2024). Enhancing Knowledge Sharing Processes via Automated Software Documentation Management Systems Using Gen AI Software Tools, 2024 XXXIV International Scientific Symposium Metrology and Metrology Assurance (MMA), Sozopol, Bulgaria, 2024, pp. 1-6, doi: 10.1109/MMA62616.2024.10817681
Goel, P. K., Verma, S., Mahajan, V., Agarwal, V., Gupta, A., & Ashok, A. (2024). Cloud data security and privacy: measures and attacks for better security and privacy (pp. 115-133). https://doi.org/10.58532/v3bfit2p3ch2 H dd d A T h i B N i M H d i K & B h i G 201 P d i d l f k l d
Haddadpoor, A., Taheri, B., Nasri, M., Heydari, K., & Bahrami, G. (2015). Process documentation: a model for knowledge management in organizations. 27(5), 347-350. https://doi.org/10.5455/MSM.2015.27.347-350
Königstorfer F. & Thalmann S. (2022). AI Documentation: A path to accountability, Journal of Responsible Technology, Volume 11, 2022, 100043, ISSN 2666-6596, https://doi.org/10.1016/j.jrt.2022.100043
Lim, S.L.O., Lim, H.M., Tan, E.K., Tan, TP. (2020). Examining Machine Learning Techniques in Business News Headline Sentiment Analysis. In: Alfred, R., Lim, Y., Haviluddin, H., On, C. (eds) Computational Science and Technology. Lecture Notes in Electrical Engineering, vol 603. Springer, Singapore, https://doi.org/10.1007/978-981-15-0058-9_35
Mahmood, A., Okumus, I. (2017). Design and Implementation of an Electronic Document Management System. Mehmet Akif Ersoy Üniversitesi Uygulamalı Bilimler Dergisi, 1(1), 9-17. Year 2017 Volume: 1 Issue: 1
Navidi, F., Hassanzadeh, M., Zolghadr Shojai, A. (2019). Organizational knowledge documentation in project-based institutes. https://www.emerald.com/insight/content/doi/10.1108/EL-10-2015-0196/full/html
Nilekar, P. (2024). Documentation Management System with PDF Chat. International Journal for Research in Applied Science and Engineering Technology. https://doi.org/10.22214/IJRASET.2024.62303
Regla, A.I. & Marquez, P.S. (2020). Workplace Document Management System Employing Cloud Computing and Social Technology. In: Alfred, R., Lim, Y., Haviluddin, H., On, C. (eds) Computational Science and Technology. Lecture Notes in Electrical Engineering vol 603. Springer Singapore. https://doi.org/10.1007/978-981-15-0058-9 40
Sebastian, G. (2023). Privacy and Data Protection in ChatGPT and Other AI Chatbots: Strategies for Securing User _ Information. Social Science Research Network. https://doi.org/10.2139/ssrn.4454761
Tin, T., Seow, X., Wong, E., Lee, K. T., & Ali, A. (2024). Interactive chatbot for PDF content conversation using an LLM language model. International Journal of Advanced Computer Science and Applications, 15(9). https://doi.org/10.14569/IJACSA.2024.01509105
Copyright Academic Conferences International Limited 2025