Learning Probabilistic Relational Models Using an Ontology of Transformation Processes

Munch, Mélanie; Wuillemin, Pierre-Henri; Manfredotti, Cristina; Dibie-Barthelemy, Juliette; Dervaux, Stéphane

doi:10.1007/978-3-319-69459-7_14

Cited by 6 publications

(4 citation statements)

References 8 publications

(8 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Orienting arcs occurs after the baseline network has been established. This is a data-driven methodology, which contrasts with existing methodologies to create CBNs directly from ontologies [58], [59], [60] instead of from data.…”

Section: F Authoritative Medical Ontologies (Amos): Icd-10-cmmentioning

confidence: 99%

A Custom GPT Health Recommender System Informed by Causal Bayesian Networks and Authoritative Ontologies (Preprint)

Hu,

Kerschberg

2024

Preprint

View full text Add to dashboard Cite

BACKGROUND ChatGPT and other large language models (LLMs) are trained on extensive text data; they learn patterns and associations within the training text without an inherent understanding of underlying causal mechanisms. Establishing causation necessitates controlled experiments and observations, and as of May 2024, ChatGPT lacks access to experimental data and the capacity to learn analytical models from data. Recent advancements from OpenAI enable the creation of custom Generative Pre-trained Transformers (GPT) models using their GPT Builder. These custom GPTs can be tailored with causal knowledge from Causal Bayesian Networks (CBN), thereby producing a knowledgeable, health-recommender system with causal expertise. This system is not only easily accessible for patients, clinicians and other users, but also has the potential to significantly improve healthcare outcomes. OBJECTIVE This paper presents a practical solution –a custom GPT model integrated with Causal Bayesian Networks (CBNs) informed by Authoritative Medical Ontologies (AMOs) as prior foundational knowledge. AMOs are robust biomedical ontologies that encapsulate the expert knowledge of their creators. By utilizing structured information contained within these ontologies, we can generate an informed CBN which can be used to improve a custom GPT as a health recommender system. These enhanced GPTs offer profound insights into cause-and-effect among co-morbid symptoms within the disease domain. METHODS To demonstrate our recommender system, we learn a CBN using NIMH data for patients of Alzheimer’s Disease and augment this with causal knowledge from the International Classification of Diseases Version 10 Clinical Modification (ICD-10-CM). We compute the CBN using the Max-Min Hill-Climbing (MMHC) algorithm. We generate two separate CBNs using MMHC to compare predictive accuracies between a baseline and a CBN modified with causal mechanisms from ICD-10-CM. Our previous research using this method has resulted in a modified CBN that reflects the causal claims in the AMOs and agrees with both the AMOs and the observational dataset. With this causal model, we build a custom GPT using OpenAI’s GPT Builder. RESULTS The custom GPT contains both potentially causal and correlations among symptoms, as well as conditional probabilities for these relationships. The GPT will also contain knowledge of causal mechanisms from ICD-10-CM, extended information regarding symptom variables, and references to existing literature regarding comorbid Alzheimer’s Disease symptoms. Furthermore, because the modified CBN model establishes potentially causal relationships among symptoms which can be verified in existing epidemiological research, we can verify that the custom GPT also establishes these causal relationships. This creates a GPT that agrees with the modified CBN, which is a representation of existing subject matter expertise in the disease domain. CONCLUSIONS To obtain our CBN, we’ve used a previous methodology which obtains ordered variable pairs from authoritative ontology ICD-10-CM as prior expertise. In our past research, a CBN that is learned using MMHC can be improved significantly by considering prior sources of knowledge, if the algorithm is modified appropriately. This source of prior knowledge can be validated in existing literature as a sequence of events in AD progression, specific causal mechanisms among comorbid symptoms, and conditional probabilities from a Bayesian Network. The resulting modified network provides insight into the causal relationships expressed in the AD data and takes advantage of the expertise and knowledge contained in the AMOs. Since inferring causality from a CBN does not exist in a vacuum, the relationships within the CBN, regardless of the strength of the conditional probabilities, must be explored further. A custom GPT extends the causal knowledge within a CBN with general knowledge of symptoms and diseases, providing a tool capable of suggesting causal inference based on the analysis of real patient data. With a LLM, uncertainty is present but reasoned with, as it is present but reasoned with in the prior, posterior, and likelihood in Bayes Theorem. Moving forward, we would like to explore using ChatGPT to produce ordered variable pairs and to automate the validation of potentially causal information.

show abstract

Section: F Authoritative Medical Ontologies (Amos): Icd-10-cmmentioning

confidence: 99%

A Custom GPT Health Recommender System Informed by Causal Bayesian Networks and Authoritative Ontologies (Preprint)

Hu,

Kerschberg

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…In a more recent study, PRMs learned from data and the semantic information of the ontology (Munch et al 2017) before building BNs. Munch et al (2019) developed a method to identify new causal relationships from an ontology and (Munch et al 2021) use this framework for decision making in the field of food packaging composite design.…”

Section: Related Studies and The Utility For Microfiltration Plantsmentioning

confidence: 99%

Decision Support Tool for the Agri-Food Sector Using Data Annotated by Ontology and Bayesian Network

Baudrit

Buche

Leconte

et al. 2022

International Journal of Agricultural and Environmental Information Systems

View full text Add to dashboard Cite

The scientific literature is a valuable source of information for developing predictive models to design decision support systems. However, scientific data are heterogeneously structured expressed using different vocabularies. This study developed a generic workflow that combines ontology, databases and computer calculation tools based on the theory of belief functions and Bayesian networks. The ontology paradigm is used to help integrate data from heterogeneous sources. Bayesian network is estimated using the integrated data taking into account their reliability. The proposed method is unique in the sense that it proposes an annotation and reasoning tool dedicated to systematic analysis of the literature, which takes into account expert knowledge of the domain at several levels: ontology definition, reliability criteria and dependence relations between variables in the BN. The workflow is assessed successfully by applying it to a complex food engineering process: skimmed milk microfiltration. It represents an original contribution to the state of the art in this application domain.

show abstract

“…Once the relational schema is defined, the PRM can be learned using the database B extracted from the knowledge base [8]. This PRM can then be instantiated in order to obtain a BN representing our learned model.…”

Section: From Stack Models To Prmsmentioning

confidence: 99%

Identifying Control Parameters in Cheese Fabrication Process Using Precedence Constraints

et al. 2018

Self Cite

View full text Add to dashboard Cite

Modeling cheese fabrication process helps experts to check their assumption on the domain such as finding which parameters (denoted as control parameters) can explain the final products and its properties. This modeling is however complex as it involves various parameters and a reasoning over different steps. Our previous work presents a method to learn a probabilistic relational model in order to check a user's (an expert on the considered domain) assumption on a transformation process domain, using a knowledge base of this domain and his expert knowledge. However this method did not include temporal information, and thus the learned model is not enough to reason on the cheese fabrication process. In this article we present an extension of our previous work that allows a user to integrate causal and temporal information represented by precedence constraints in order to model a cheese fabrication process. This allows the user to check his assumption to identify the transformation process control parameters.

show abstract

Learning Probabilistic Relational Models Using an Ontology of Transformation Processes

Cited by 6 publications

References 8 publications

A Custom GPT Health Recommender System Informed by Causal Bayesian Networks and Authoritative Ontologies (Preprint)

A Custom GPT Health Recommender System Informed by Causal Bayesian Networks and Authoritative Ontologies (Preprint)

Decision Support Tool for the Agri-Food Sector Using Data Annotated by Ontology and Bayesian Network

Identifying Control Parameters in Cheese Fabrication Process Using Precedence Constraints

Contact Info

Product

Resources

About