Pharmaceutical chatbot assistant using generative AI and knowledge graph with specific pharmaceutical resources

Generative AI has changed the path of finding for information and solving problems quite easier. Instead of just giving links, it understands what we want, explains things effectively, and even assist us think automatically. Large Language Models like GPT, LLaMA, and PaLM are widely used for creating human-like text, but they still face some limitations such as making incorrect facts, giving unreliable answers, failing to follow proper reasoning steps and more. These limitations become more intense in situations where accurate and reliable information is necessary. To overcome this, Retrieval Augmented Generation was introduced to provide the most efficient and reliable answers. During the response process, it makes the models to find the right information from external sources. Even though this is deficient for model to give a true and meaningful answer. So, Knowledge Graphs are used here to improve its understanding level even better and to improve its ability to get ideas from various steps. This study gives an in-depth review and design of a system that combines both Retrieval Augmented Generation and Knowledge Graphs to deliver right answers. This tirelessly examines the mistakes that take place when language models work alone. The retrieval improves as it gives a verified information from trusted sources and give it to the AI while responding for answers. So, it acts like instead of guessing from memory, the AI uses real facts comparison to make the answers so accurate while they responding for output. Results from both experimental and analysis acknowledges that using Retrieval-Augmentation Generation with Knowledge Graphs works much effectively than the normal models which run on their own. This combination approach reduces mistakes and give accurate and trustworthy results when compared with models that run alone.

The convergence of Generative AI and agentic frameworks is fundamentally transforming enterprise data governance, offering a solution to the exponential growth in data complexity that traditional manual methods cannot address. This emerging paradigm shift introduces the concept of a Data Stewardship Co-Pilot—an intelligent, AI-powered partner that guides organizations through comprehensive data governance while democratizing access to complex data environments. The evolution from manual processes to intelligent automation employs specialized horizontal and vertical AI agents that autonomously handle diverse governance tasks with unprecedented precision and speed. These agents work within a sophisticated architecture comprising prompt stores, intuitive interfaces, vector-based indexing, context search engines, agent pools, knowledge graphs, and feedback mechanisms. By merging Generative AI's conversational capabilities with agentic frameworks' analytical depth, organizations can achieve significant improvements in efficiency, data quality, scalability, and collaboration while reducing operational costs and enhancing compliance standards. Leading technology platforms are actively facilitating these innovations, positioning AI-assisted data stewardship as a strategic differentiator in the increasingly data-driven business landscape.

This paper addresses the deficiencies in traditional design methods for exhibition spaces, particularly in terms of efficiency, customization, iterative speed, and design cycles. Specifically, it highlights the issues with the emerging digitalized design methods based on Generative Artificial Intelligence (genAI), such as non-standard and unstructured input expressions, high inconsistency in output results, and low repeatability. By utilizing the Knowledge Graph (KG) to achieve accurate semantic representation and orderly contextual association, as well as structured and standardized expression, this study explores and establishes a KG-genAI-based paradigm for the digital and intelligent design of exhibition spaces (KG-genAI-PDIDES). This paradigm drives Midjourney and Stable through unstructured natural language prompting methods for the digital design of UAV exhibition spaces. Comparative analysis of the results demonstrates that the proposed KG-genAI-PDIDES significantly enhances the standardization of input expression and consistency of output results, offering a new design paradigm for the digitization of exhibition spaces.

The paper reviews, represents and captures knowledge about industrial water treatment processes and predicative models, in order to facilitate management of relevant knowledge. The proposed approach is based on a Knowledge Graph (KG), which integrates process knowledge and predictive models, adding context to their application and usage; improves problem and data understanding by facilitating communication between data analysts and process engineers, providing clear, human-readable explanations; and facilitates answering process-related knowledge questions and provides answers that include relevant data elements, models, and key performance indicators (KPIs).Further, the paper includes examples of how the KG can be used in practice. Directions and recommendations are provided, as well as research guidelines of how the KG can augment generative AI approaches, paving the way for the development of retrieval-augmented knowledge management models and systems.

EHR systems has transformed healthcare data management but challenges related to interoperability, cost and security remain significant. Role of ontologies in EHR for representing the unstructured and semi-structured data found in different medical records. such as doctor's prescription, diagnostic report, and other treatment-related data and different biomedical knowledge bases are well established. The ontologies and knowledge graphs are used for representing and structuring the unstructured data so that it can be used for coding and retrieving the relevant data in knowledge bases. Knowledge Graphs are flexible and adaptable as it organizes data into interconnected structures that represent relationships between entities such as disease, treatments, and progressions. Implied relationship between many diseases is known however they involve complex interdependencies. Ontology merging plays an important role in showing the causal relationship among such diseases. The objective of ontology merging is to integrate pertinent features of ontologies, like axioms, persons, and annotations, into the resultant ontologies that can be used for better establishing the semantic relationship between two diseases. In this study two diseases considered are Hepatitis and Dyslipidemia and converging their ontologies a HypDysLip ontology is created. The created ontology is further enriched using the generative AI tool ChatGPT.

In recent years, the landscape of digital biomedicine and healthcare has been reshaped due to the disruptive breakthroughs in AI-facilitated by tremendous data and high-performance computers, large language models (LLMs) have transformed information technology from accessing data to performing analytical tasks. While demonstrating unprecedented capabilities, LLMs have been found unreliable in tasks requiring factual knowledge and rigorous reasoning. Biomedicine and healthcare, as an important vertical domain rapidly benefitting from progress in AI, necessitates strict requirements on the accuracy, controllability, and interpretability of analytical models, posing critical challenges for LLMs. Despite recent studies addressing the hallucination problem of LLMs, research on empowering LLMs with the ability to plan, reason, and ground with explicit knowledge has also started to prosper, especially in the biomedicine and healthcare domain. On the other hand, biomedical data are enormous and notoriously complex, coming from various sources (e.g., biomedical knowledge bases, online literature, and hospitals) and bearing various modalities (e.g., tables, texts, images and time-series). Healthcare professionals have spent decades collecting, cleaning, and curating various types of data. The processes are extremely costly, producing various datasets with different data schemas, coding systems, and quality standards, many privately owned by the creators, making their integrative analysis and utilization through unified AI techniques still rather challenging. The generalizability of LLMs across different types of data endow them strong promises in automating the processing of large-scale complex healthcare data such as into unified knowledge graphs (KGs). Our goal in this survey is to systematically investigate and summarize recent studies on the unification of LLMs and KGs, towards fully utilizing the value of complex data, unleashing the power of generative AI, and expediting next-generation AI for biomedicine and healthcare applications.

External knowledge graphs (KGs) can be used to augment large language models (LLMs), while simultaneously providing an explainable knowledge base of facts that can be inspected by a human. This approach may be particularly valuable in domains where explainability is critical, like human trafficking data analysis. However, creating KGs can pose challenges. KGs parsed from documents may comprise explicit connections (those directly stated by a document) but miss implicit connections (those obvious to a human although not directly stated). To address these challenges, this preliminary research introduces the GAME-KG framework, standing for “Gaming for Augmenting Metadata and Enhancing Knowledge Graphs.” GAME-KG is a federated approach to modifying explicit as well as implicit connections in KGs by using crowdsourced feedback collected through video games. GAME-KG is shown through two demonstrations: a Unity test scenario from Dark Shadows, a video game that collects feedback on KGs parsed from US Department of Justice (DOJ) Press Releases on human trafficking, and a following experiment where OpenAI’s GPT-4 is prompted to answer questions based on a modified and unmodified KG. Initial results suggest that GAME-KG can be an effective framework for enhancing KGs while simultaneously providing an explainable set of structured facts verified by humans.

This paper presents a semantic framework to bridge the gap between IT-OT integration in industrial environments. The proposed solution addresses fundamental challenges of PHM (prognostics and health management) by providing contextualized semantic information from the shop floor to enterprise IT systems. Built upon an OPCUA (Open Platform Communications Unified Architecture) aggregation server architecture, the framework leverages OPCUA Information Models and companion specifications as its foundation for semantic representation. By transforming these models into knowledge graphs stored in RDF format, the system enables sophisticated semantic information retrieval through SPARQL-based semantic queries that can traverse complex relationships between equipment, processes, and operational parameters. The framework further implements GraphQL to automatically generate a Type schema derived from OPCUA types, creating a unified query interface that facilitates IT-like interaction with industrial data. This semantic approach significantly improves fault diagnostics, predictive maintenance, and anomaly detection by preserving contextual relationships that are often lost in traditional data integration methods. Furthermore, the GraphQL schema provides a structured foundation for generative AI applications to formulate contextually appropriate queries, extract relevant maintenance insights, and generate human-interpretable explanations of equipment health patterns, all while maintaining semantic fidelity across the IT-OT boundary. The vertical integration capability ensures that domain-specific models remain coherent across organizational levels such as line, area, floor, etc., enabling PHM practitioners to implement more effective condition-based maintenance strategies with improved visibility into causal factors affecting equipment reliability and performance.

Modern enterprise service desks are burdened with high incident volumes, inconsistent triage accuracy, and extended resolution times. This paper introduces a novel framework combining generative AI with dynamic knowledge graphs to streamline incident triage in IT service management (ITSM) environments. The proposed system leverages real-time graph-based context retrieval from CMDBs, historical tickets, and dependency maps to enable LLM-driven agents to understand, classify, and route incidents with minimal human intervention. By integrating graph embeddings with generative prompting strategies, the assistant adapts to evolving service environments and improves root-cause correlation. Evaluations on real-world ITSM datasets demonstrate a 73% reduction in triage time, a 41% increase in first-call resolution accuracy, and significant gains in knowledge reuse. The results underline the potential of agentic knowledge-graph architectures to transform reactive service desks into proactive resolution engines.

The integration of artificial intelligence into data engineering processes represents a paradigmatic shift in how organizations manage, process, and derive value from their data assets. This comprehensive technical review examines the transformative impact of machine learning on traditional Extract, Transform, Load (ETL) workflows and data pipelines. Starting with intelligent data extraction capabilities that leverage natural language processing and computer vision, continuing through adaptive transformation logic and smart loading optimization, AI enhances every aspect of the data engineering lifecycle. Advanced anomaly detection and automated quality control mechanisms enable proactive identification of issues before they impact downstream systems. Reinforcement learning algorithms optimize resource allocation while self-tuning pipelines continuously refine operational parameters without human intervention. Despite significant benefits, organizations face substantial implementation challenges including explainability limitations, skills gaps, legacy system integration, and governance considerations. The emerging landscape features knowledge graphs for semantic understanding, generative AI for pipeline creation, and cross-organizational data fabrics with embedded intelligence innovations that collectively blur traditional boundaries between data engineering and data science disciplines.

With Generative AI (GenAI) technologies increasingly incorporated into enterprise solutions—ranging from automated reports to interactive business analytics—the demand for effective monitoring and semantic validation frameworks has expanded exponentially. This overview examines the convergence of GenAI with Knowledge Graphs (KGs), particularly how AI-based monitoring can improve model reliability, minimize hallucinations, and offer interpretability in real-time use cases. Through integration of recent research findings, benchmarking studies, and industrial case applications, the paper brings into focus the revolutionizing capability of combining structured ontologies with neural networks. The review ends with a future research and practice roadmap to make GenAI scalable, secure, and aligned to society.

QCI Robot with Generative AI Knowledge Graph for Taiwanese/Japanese Co-Learning Model Application

Enhancing foundation models for scientific discovery via multimodal knowledge graph representations