Jul 2026· International Journal of Computer Science & Information System· Vol 11, pp. 25-42· 0 citations
TL;DR
The findings suggest that future literature analysis platforms will increasingly integrate Generative AI with retrieval systems, human-in-the-loop verification, autonomous research agents, and standardized governance frameworks to produce trustworthy, scalable, and transparent scientific intelligence.
Abstract
The rapid expansion of scientific publications across disciplines has made traditional literature review methodologies increasingly difficult to execute efficiently. Researchers must analyze thousands of articles, identify emerging trends, synthesize evidence, detect research gaps, and evaluate methodological quality within limited timeframes. Generative Artificial Intelligence (Generative AI), powered by large language models, transformer architectures, retrieval-augmented generation, and intelligent knowledge representation techniques, has emerged as a transformative solution for automated scientific literature analysis. Unlike conventional text mining approaches that primarily perform keyword matching or statistical extraction, Generative AI demonstrates contextual understanding, semantic reasoning, automated summarization, question answering, citation synthesis, hypothesis generation, and research trend identification. These capabilities significantly improve the efficiency, scalability, and quality of scientific knowledge management while reducing researcher workload.
This review systematically examines recent developments in Generative AI for automated scientific literature analysis by synthesizing evidence from the provided contemporary literature covering artificial intelligence, workflow automation, cloud intelligence, cybersecurity, financial AI, process mining, enterprise automation, reinforcement learning, digital transformation, and intelligent computing infrastructures. The review develops a comprehensive analytical framework describing the complete literature-analysis pipeline, including literature acquisition, document preprocessing, semantic embedding, knowledge extraction, contextual reasoning, automated synthesis, evidence validation, and research recommendation generation. Furthermore, the study critically evaluates technological enablers such as transformer-based architectures, cloud-edge computing infrastructures, retrieval-augmented generation, agentic AI, workflow automation, and scalable enterprise AI systems that collectively support intelligent literature analysis (Krishnan & Bhat, 2025; Kumar, 2025; Venkiteela, 2026).
The review identifies significant opportunities in accelerating systematic reviews, improving interdisciplinary knowledge discovery, reducing information overload, supporting evidence-based decision making, and enabling continuous scientific monitoring. Simultaneously, important challenges remain concerning hallucination, citation reliability, explainability, reproducibility, privacy, governance, computational scalability, and ethical deployment. The findings suggest that future literature analysis platforms will increasingly integrate Generative AI with retrieval systems, human-in-the-loop verification, autonomous research agents, and standardized governance frameworks to produce trustworthy, scalable, and transparent scientific intelligence. This review contributes a structured conceptual framework that integrates recent advances in Generative AI with automated scientific literature analysis while identifying future research opportunities for developing reliable AI-assisted scientific discovery ecosystems.
Scientific publishing, digital repositories, patents, and multidisciplinary research datasets have expanded rapidly, making traditional literature review methods increasingly inefficient. Large Language Models (LLMs) address this challenge by enabling intelligent knowledge discovery, semantic search, literature summarization, research gap identification, hypothesis generation, citation assistance, and academic writing support. By integrating Retrieval-Augmented Generation (RAG), vector databases, knowledge graphs, citation networks, and domain-specific ontologies, LLMs improve contextual relevance, reduce hallucinations, and enhance research accuracy. These capabilities accelerate interdisciplinary collaboration, automate research workflows, and support evidence-based decision-making. However, challenges such as hallucination, bias, outdated knowledge, explainability, privacy, intellectual property, reproducibility, and computational requirements remain significant. Modern AI-assisted research systems increasingly incorporate human-in-the-loop validation, explainable AI, and responsible governance to ensure trustworthy outcomes. This study presents a conceptual framework that combines semantic retrieval, intelligent reasoning, automated literature analysis, and workflow orchestration, demonstrating how LLM-powered systems can transform scientific research into scalable, accurate, ethical, and collaborative knowledge discovery processes.
Narendra Karmarkar, Iyengar P.K· International Journal of Eme...· 0 citations
Generative Artificial Intelligence (AI) has become a rapidly expanding area of scientific research, generating a growing body of literature across technical and applied domains. This study provides a comprehensive bibliometric analysis of global generative AI research to characterize its publication growth, disciplinary and geographical distribution, institutional participation, funding patterns, citation performance, and thematic development. The analysis covers 22,758 Scopus-indexed journal articles and conference papers published between 2020 and 2025, identified using the phrase “generative artificial intelligence” enclosed in double quotation marks in TITLE-ABS-KEY fields. A reproducible computational workflow was used to examine publication output, document types, subject areas, countries, institutions, funding sponsors, citation patterns, and keyword co-occurrence. Citation analysis incorporated annualized citation rates and cohort-normalized annual citation rates to improve comparisons across publication years. Results show a pronounced acceleration in publication output after 2022, with an approximate 105% compound annual growth rate over 2020–2025. Computer Science remained the largest subject area, while substantial representation extended across Engineering, Social Sciences, Medicine, Mathematics, and other domains. Publication activity was concentrated among leading countries and institutions, with the United States and China recording the highest output. Funding analysis identified major national and international sponsors through publication–sponsor associations. Citation performance varied substantially across cohorts, with the 2023 cohort exhibiting the highest cohort-normalized annual citation rate (1.58). Keyword analysis revealed a thematic shift from early AI and GAN-related research toward generative AI and large language model themes, alongside education, innovation, human–AI interaction, and responsible AI. The findings provide an evidence-based, multidimensional characterization of the rapidly evolving generative AI research landscape.
Sofia Stamou, Matina Kiourexidou· Information· 0 citations
The exponential growth of scientific publications creates a critical challenge for researchers attempting to navigate their fields. Manual literature reviews, once sufficient for identifying research opportunities, now consume disproportionate time and often lack comprehensiveness. This paper presents an automated research gap detection system that integrates natural language processing, citation network analysis, and ensemble machine learning to identify research gaps across scientific literature systematically. The proposed system uses transformer-based models (SciBERT, BioBERT) for semantic understanding, graph neural networks for citation structure analysis, and Support Vector Machines, Random Forests, and Gradient Boosting for gap classification. We implement a complete pipeline processing documents at scale, extracting semantic content, analyzing citation relationships, and identifying knowledge gaps through multiple complementary techniques. The system was designed to detect three categories of gaps: knowledge discrepancies (conflicting information), knowledge voids (completely missing information), and methodological limitations (inadequate research methods). Experimental evaluation across multiple scientific domains demonstrates that the system identifies research opportunities with accuracy comparable to expert assessments while processing millions of documents efficiently. The results show 78% accuracy in predicting emerging research areas up to two years in advance and 88% validation rate for identified gaps when reviewed by domain experts.
Adesh V. Patil, Somanath J. Salunkhe· Dandao Xuebao/Journal of Bal...· 0 citations
Nowadays, Generative Artificial Intelligence (GenAI) is making a significant impact on research activities and library services. In order to identify the core dimensions of contemporary research on GenAI applications in developing Research Support Services (RSS) at academic libraries a bibliometric analysis of publications was conducted. Data were collected from the Web of Science database, comprising 362 articles published during the period 2016–2025. Together, the bibliometric analysis with Microsoft Excel, VOSviewer, Carrot2, and CiteSpace revealed publication trends, international collaborations, keyword co-occurrence, topic clustering, and temporal research evolution. Results indicate rapidly increasing publication output, particularly from 2023 onward, reflecting growing interest from the academic community. The USA, the UK, and India play central roles in the research collaboration network. Prominent themes focus on GenAI applications in intelligent reference services, chatbots, recommendation systems, knowledge discovery, research data management, and AI literacy training, alongside emerging issues such as ethics and data security. This study provides a scientific foundation for directing RSS development at academic libraries in the GenAI context.
Phan Truong Nhat, Hong Sinh Nguyen· IAFOR Journal of Literature...· 0 citations
In this systematic literature review, we aimed to identify and thoroughly analyze the existing scientific knowledge related to a notably emerging field of generative AI. In strict adherence to the SPAR-4-SLR protocol, we focus on 1,104 peer-reviewed articles from the Scopus database, published between 2015 and 2025. Using bibliometric and thematic mapping methods, we address two main research questions: the first concerns the major publication trends in GenAI research, and the second deals with the intellectual structures and thematic domains shaping the field. Our results indicate an abrupt increase in scientific production since 2022, a consequence of the launch of models such as GPT, DALL·E, and Stable Diffusion, which are becoming increasingly powerful. We further identify five main research clusters: the technological foundations of generative AI, where researchers focus on building and utilizing LLMs and deep learning; professional and educational applications; ethical and governance issues; AI-assisted creativity; and user perceptions. Additionally, we find that higher education plays a significant role in the area, both in the application of ideas and the exploration of relevant questions. This review highlights the field's strong interdisciplinary character and, at the same time, reveals the current challenges that the sector faces. We outline a systematic research program to guide further studies of the implementation, impact, and problems of GenAI in enterprises and communities.
Majdouline Attaoui, Wissal Attaoui, Anas Moukrim et al.· International journal of mul...· 0 citations
Artificial intelligence (AI), particularly generative AI and large language models (LLMs), is rapidly transforming research planning by enhancing literature synthesis, research problem formulation, methodology selection, and workflow management. Despite this rapid adoption, limited evidence exists regarding the intellectual structure, thematic evolution, and responsible integration of AI within research planning. Therefore, this study systematically maps the scientific landscape of AI-assisted research planning through a bibliometric analysis of publications indexed in the Scopus database between 2017 and 2026. Bibliographic data were analysed using Biblioshiny (Bibliometrix) and VOSviewer to examine publication trends, influential contributors, collaboration networks, intellectual structure, and thematic development. The findings reveal a rapidly expanding research domain, with an annual publication growth rate of 51.08%, reflecting the increasing integration of AI into research planning. Science mapping identifies research management, artificial intelligence, machine learning, and large language models as the dominant research themes, while thematic evolution demonstrates a shift from foundational AI technologies towards AI-enabled research management and workflows. However, the analysis also reveals a significant conceptual imbalance, with technological advancements outpacing the development of governance, trust, ethics, and human–AI collaboration. To address this gap, the study proposes a five-phase framework comprising Strategic Conceptualisation, Systematic Literature Synthesis, Methodology Selection, Governance, Ethics and Trust, and Continuous Reflection and Feedback, positioning the human touch, curiosity, critical thinking, creativity, contextual expertise, and ethical judgement, as the foundation of responsible AI-assisted research planning. The study contributes both a comprehensive knowledge map and a human-centred conceptual framework that provide practical guidance for researchers, scholars, and policymakers seeking to promote the responsible, transparent, and sustainable integration of AI throughout the research planning process.
Niranjan Devkota, M. Siddique, Dipendra Karki et al.· International Research Journ...· 0 citations