Structured Extraction of Real World Medical Knowledge using LLMs for Summarization and Search

A web server for automatic analysis and extraction of relevant biological knowledge

Social media has extensive content volunteered by drug users, who also mentions drug side-effects. Such mentions may indicate side-effects that are heretofore unknown to the medical professionals and researchers, pharmaceutical companies, and regulators. To study and monitor the drug and its potential symptoms and side-effects for epidemiological surveillance of drug trends, the research needed a higher level of knowledge extraction and improvement in domain knowledge. Our aim is to methodically classify the tweets that mention drug side-effects, and by applying our model we improved the medical knowledge extraction from our data-set. We crawled 7 million tweets related to marijuana concentrate and cannabis (known as marijuana) the tweet text mentioned about marijuana usage, marijuana consumption, the medicinal value of marijuana and symptoms or side-effects of marijuana. It is essential to identify the drug side-effects in the tweets in order to create a bag of words relating to each category of marijuana. This will assist us in the classification of tweets under different categories of marijuana-based on side-effects. We manually created multi-domain knowledge, modeled in DsOn(Drug and Symptom Otology), to facilitate the extraction of semantic parsing from people opinion on Social Media, through a combination of medical taxonomy dictionary, and semantics relationship based techniques from UMLS, PubMed, and DBpedia. Experiment results showed an improvement in recall from 0.14 (without ontology model) to 0.23 (with ontology model).

Most enterprises, technology and otherwise, are routinely collecting massive amounts of unstructured information from customer interactions and then attempting to extract knowledge from this information in order to improve core activities such as product development, customer support, and marketing. The knowledge engineering processes for extracting knowledge from this information are often largely manual and extremely inefficient in both cost and time. Therefore, software automation of these manual activities through the creation of highly user-centric Knowledge Engineering Software Products (KESPs) is critical to enabling the rapid and efficient extraction of high-quality knowledge. The primary intent of this paper is to provide a comprehensive theory, including its application and implementation, for developing high-value Knowledge Engineering Software Products. To this end, we have created a representation-based approach to the design and development of KESPs. The theoretical framework of our representational-based approach is the integrated metarepresentational model (IMRM) which provides a natural sequence of representations for guiding the development of complex artifacts such as KESPs. The application of the IMRM to the development of high-value KESPs resulted in the integrated representation-based process methodology (IRPM) which combines, in a rational and structured manner, methods and tools from the technical domains of Knowledge Engineering, Product Design, and Software Engineering. Each domain contributes a distinct set of methods to the IRPM. The knowledge engineering domain provides tools—such as the CommonKADS Agent/Task model—for modeling current work processes that the KESP will automate. The product design domain provides formal tools—such as the House of Quality, Function Structure, Morphological Matrix, and Utility Function—for explicitly defining the user needs for the KESP, and for exploring different design concepts in order to ensure the KESP is high-quality and low-cost. The software engineering domain provides tools—such as Unified Modeling Language (UML) Use Case, Component, and Class diagrams—in order to ensure that a reliable and easy to use KESP is delivered on time and within budget. We have demonstrated the feasibility of the IRPM by implementing it within the context of a real knowledge engineering problem involving the extraction of problem-solution pairs from customer service requests in order to create “smart” products and services. The developed KESP, called the “Service Request Portal” (SRP), used search and content filters to achieve a 30% productivity improvement over the previously manual work process.

Nursing students' experiences with the objective structured clinical examination (OSCE): A qualitative study

This research aimed to extract medical diagnostic knowledge about sleep apnea by applying theories from process control management, library science, and knowledge management.We interviewed the President of the International Sleep Science Technology Association (ISSTA), a medical doctor, on the subject of sleep apnea, and validated the research findings with four other sleep apnea experts to achieve the following: A formal knowledge extraction procEduccre was established for sleep apnea.All medical knowledge pertaining to sleep apnea was mapped out.

Genomic data mining and knowledge extraction is an important problem in bioinformatics. Some research work has been done on unknown genome identification and is based on exact pattern matching of n-grams. In most of the real world biological problems exact matching may not give desired results and the problem in using n-grams is exponential explosion. In this paper we propose a method for genome data classification based on approximate matching. The algorithm works by selecting random samples from the genome database. Tolerance is allowed by generating candidates of varied length to query from these sample sequences. The Levenshtein distance is then checked for each candidate and whether they are k-fuzzily equal. The total number of fuzzy matches for each sequence is then calculated. This is then classified using the data mining techniques namely, naive Bayes, support vector machine, back propagation and also by nearest neighbor. Experiment results are provided for different tolerance levels and they show that accuracy increases as tolerance does. We also show the effect of sampling size on the classification accuracy and it was observed that classification accuracy increases with sampling size. Genome data of two species namely Yeast and E. coli are used to verify proposed method.

The amount of data being generated and stored is growing exponentially, owed in part to the continuing advances in computer technology. These data present tremendous opportunities in data mining, a burgeoning field in computer science that focuses on the development of methods that can extract knowledge from data. In many real world problems, these data mining algorithms have access to massive amounts of data. Mining all the available data is prohibitive due to computational (time and memory) constraints. Much of the current research is concerned with scaling up data mining algorithms (i.e. improving on existing data mining algorithms for larger datasets). An alternative approach is to scale down the data. Thus, determining a smallest sufficient training set size that obtains the same accuracy as the entire available dataset remains an important research question. Our research focuses on selecting how many (sampling) instances to present to the data mining algorithm. The goals of this paper is to study and characterize the properties of learning curves, integrate them with Chebyshev Bound to come up with an efficient general purpose adaptive sampling schedule, and to empirically validate our algorithm for scaling down the data.

IoT deployments generate data of the real world in an automated fashion without direct user involvement. With increasing scale of these IoT deployments the extraction of the right knowledge about the real world from a vast amount of IoT data and efficient decisions is a challenging endeavor. While solutions to deal with large amounts of IoT data are slowly emerging, potential users of IoT solutions, or policy makers find it difficult to assess the actual usefulness of investing in IoT deployments or selecting adequate deployment strategies for a particular business domain. Despite the recent hype generated by consultancy companies and IoT vendors about the IoT, there is still a lack of experience in assessing the utility and benefits of IoT deployments as many IoT deployments are still in their early stages. In order to capture such experience quicker and derive best practices for IoT deployments, systematic tools, or methodologies are required in order to allow the assessment of the goodness or usefulness of IoT deployments and a comparison between emerging IoT deployments to be performed. This chapter addresses the existing gap and proposes a novel benchmarking framework for IoT deployments. The proposed framework is complementary to the emerging tools for the analysis of big data as it allows various stakeholders to develop a deeper understanding of the surrounding IoT business ecosystem in a respective problem domain and the value proposition that the deployment of an IoT infrastructure may bring. It also allows a better decision making for policy makers for regulatory frameworks.

We propose a supervisory system that deals with actual situations and social aspects of users in ubiquitous computing environment in this paper. To realize a gentle and safe supervision, we have to consider social relationships between observers and the watched person to give adequate level of privacy to the services. We propose "uEyes", a ubiquitous supervisory system that introduces a new feature for the supervision, i.e., social context awareness. By using this feature, environmental information acquired from real world and knowledge on social activities of people are integrated effectively. From experimental results using a prototype of uEyes for supervising elderly people, we show that the social context awareness, as well as other components of the system, effectively works to support the supervision.

Legal reasoning involves case analysis in statutory as well as real world perspectives. The impact of real world perspective on case analysis poses a serious challenge to knowledge engineers for building legal expert systems. A legal expert system intends to provide intelligent support to legal professionals. The proposed legal predictive system is an attempt to predict the most probable outcome of a case according to statutory as well as real world knowledge of the legal domain. The system accepts the current fact situation of a case and analyses it interactively with legal personnel. This work introduces a frame-like knowledge structure,lattice, with two-dimensional attributes. This paper contains a detailed discussion on artificial intelligence-based case analysis of theft cases in a real world perspective.

<span>The problem is the low ability of students, and the low orientation of developing students' thinking skills in learning, especially the thinking skills metacognition which is the thinking of the 21st century skills. These factors are one of the main factors that are still oriented towards achieving knowledge in a partial-textual way and not an integrative contextual understanding of the knowledge learned. Therefore, there are efforts that can be used to resolve the problem through effective and up-to-date effective models that are effort-oriented in improving students' metacognitive thinking skills, which is through the application of critical literacy models based on project based learning approaches. The critical multiliteration model is a learning model developed based on awareness and recognition of cultural perspectives and the diversity of learning styles. In addition, the critical multiliteration model is a model that is oriented towards efforts to build students' thinking skills through multicontext, multicultural and multimedia approaches. In this context, critical multiliteration models are specifically developed as a dynamic learning approach, where students actively explore problems in real world, provide challenges, and gain deeper knowledge. This is based on the purpose of developing a model oriented to the development of metacognitive thinking skills.</span>

Entity Correspondence seeks to find instances that refer to the same real world entity. Usually, a fixed set of properties exists, for each of which the similarity score is computed to support entity correspondence. However, in a knowledge base that has properties incrementally recognized, we can no longer rely only on the belief that two instances sharing value for the same property are likely to correspond with each other: a pair of different properties that are of hierarchies or specific relations can also be evidential to corresponding instances. This paper proposes the use of second-order Markov Logic to perform entity correspondence. With second-order Markov Logic, we regard properties as variables, explicitly define and exploit relations between properties and enable interaction between entity correspondence and property relation discovery. We also prove that second-order Markov Logic can be rephrased to first-order in practice. Experiments on a real world knowledge base show promising entity correspondence results, particularly in recall.KeywordsEntity CorrespondenceSecond-Order Markov LogicProperty Relation DiscoveryKnowledge Base

To improve the hospital's ability to proactively detect infectious diseases, a knowledge-based infectious disease monitoring and decision support system was established based on real medical records and knowledge rules. The effectiveness of the system was evaluated using interrupted time series analysis. In the system, a monitoring and alert rule library for infectious diseases was generated by combining infectious disease diagnosis guidelines with literature and a real medical record knowledge map. The system was integrated with the electronic medical record system, and doctors were provided with various types of real-time warning prompts when writing medical records. The effectiveness of the system's alerts was analyzed from the perspectives of false positive rates, rule accuracy, alert effectiveness, and missed case rates using interrupted time series analysis. Over a period of 12 months, the system analyzed 4,497,091 medical records, triggering a total of 12,027 monitoring alerts. Of these, 98.43% were clinically effective, while 1.56% were invalid alerts, mainly owing to the relatively rough rules generated by the guidelines leading to several false alarms. In addition, the effectiveness of the system's alerts, distribution of diagnosis times, and reporting efficiency of doctors were analyzed. 89.26% of infectious disease cases could be confirmed and reported by doctors within 5 min of receiving the alert, and 77.6% of doctors could complete the filling of 33 items of information within 2 min, which is a reduction in time compared to the past. The timely reminders from the system reduced the rate of missed cases by doctors; the analysis using interrupted time series method showed an average reduction of 4.4037% in the missed-case rate. This study proposed a knowledge-based infectious disease decision support system based on real medical records and knowledge rules, and its effectiveness was verified. The system improved the management of infectious diseases, increased the reliability of decision-making, and reduced the rate of underreporting.

We would like to welcome you to the 11th ACM International Conference on Knowledge Capture: K-CAP 2021. Driven by the increasing demands for knowledge-based applications and the unprecedented availability of information from heterogenous sources, the study of knowledge capture is of crucial importance. Knowledge capture involves the extraction of useful knowledge from vast and diverse data sources as well as its acquisition directly from human experts. K-CAP 2021 aimed at attracting researchers from diverse areas of Artificial Intelligence, including knowledge acquisition, knowledge representation and reasoning, intelligent user interfaces, problem-solving, planning, agents, information extraction, machine learning, natural language processing, information enrichment and visualization, as well as researchers interested in cyberinfrastructures to foster the publication, retrieval and reuse of data. Nowadays knowledge is derived from an increasingly diverse set of data resources that differ with regard to their domain, format, quality, coverage, specificity, viewpoint, bias, and most importantly, consumers and producers of the data. The heterogeneity, amount and complexity of the data allow us to answer complex questions that could not be answered in isolation - requiring the interaction of different scientific fields and technologies. A goal of K-CAP is to develop such synergies using systematic and rigorous methodologies. The call for papers attracted 123 submissions from all over the world, covering a diverse range of topics spanning ontology development, information extraction, knowledge graphs, natural language processing, reasoning, entity linking, querying and knowledge-based applications. From a competitive set of high-quality submissions, we accepted 31 long research papers, 8 short papers, and 2 demo papers. We encourage everyone to attend the keynote talks that we have planned for K-CAP 2021. The highly anticipated talks by Leila Zia (Wikimedia Foundation) and Ian Horrocks (University of Oxford) will guide us to a better understanding of the future of knowledge graph and knowledge representation technologies, as they become prevalent in real world applications.

FCANN: A new approach for extraction and representation of knowledge from ANN trained via Formal Concept Analysis