Multiple Domain Knowledge Research Articles

Mapping vaccine names in clinical trials to vaccine ontology using cascaded fine-tuned domain-specific language models

BackgroundVaccines have revolutionized public health by providing protection against infectious diseases. They stimulate the immune system and generate memory cells to defend against targeted diseases. Clinical trials evaluate vaccine performance, including dosage, administration routes, and potential side effects. ClinicalTrials.gov is a valuable repository of clinical trial information, but the vaccine data in them lacks standardization, leading to challenges in automatic concept mapping, vaccine-related knowledge development, evidence-based decision-making, and vaccine surveillance.ResultsIn this study, we developed a cascaded framework that capitalized on multiple domain knowledge sources, including clinical trials, the Unified Medical Language System (UMLS), and the Vaccine Ontology (VO), to enhance the performance of domain-specific language models for automated mapping of VO from clinical trials. The Vaccine Ontology (VO) is a community-based ontology that was developed to promote vaccine data standardization, integration, and computer-assisted reasoning. Our methodology involved extracting and annotating data from various sources. We then performed pre-training on the PubMedBERT model, leading to the development of CTPubMedBERT. Subsequently, we enhanced CTPubMedBERT by incorporating SAPBERT, which was pretrained using the UMLS, resulting in CTPubMedBERT + SAPBERT. Further refinement was accomplished through fine-tuning using the Vaccine Ontology corpus and vaccine data from clinical trials, yielding the CTPubMedBERT + SAPBERT + VO model. Finally, we utilized a collection of pre-trained models, along with the weighted rule-based ensemble approach, to normalize the vaccine corpus and improve the accuracy of the process. The ranking process in concept normalization involves prioritizing and ordering potential concepts to identify the most suitable match for a given context. We conducted a ranking of the Top 10 concepts, and our experimental results demonstrate that our proposed cascaded framework consistently outperformed existing effective baselines on vaccine mapping, achieving 71.8% on top 1 candidate’s accuracy and 90.0% on top 10 candidate’s accuracy.ConclusionThis study provides a detailed insight into a cascaded framework of fine-tuned domain-specific language models improving mapping of VO from clinical trials. By effectively leveraging domain-specific information and applying weighted rule-based ensembles of different pre-trained BERT models, our framework can significantly enhance the mapping of VO from clinical trials.

Specifying features in terms of domain models: MuDForM method definition and case study

AbstractTo enable the people involved in a software development process to communicate and reason close to their area of knowledge, we are investigating and engineering a method that formalizes and integrates knowledge of multiple domains into domain models and into specifications in terms of those domain models. We follow an action research approach, starting with a diagnosis phase, in which we have previously defined a set of method objectives, and performed a systematic literature review. During action planning, we defined how we are going to develop the method—called Multi‐Domain Formalization Method (MuDForM). This paper reports on the methodical support for using a domain model as the terminology for feature specifications. During action taking, we defined an initial version of the method and set up case studies. During the evaluation phase, we performed a case study to validate how well the method helps in the specification of processes and to realize the case‐specific objectives of the customer. The case study pertains to the formalization of the ISO26262 standard for functional safety in the automotive domain. The created models are explained to the involved experts to ensure their consistency with the original text. We found that MuDForM is suitable to systematically formalize processes described in natural language, such that the resulting process models are fully expressed in terms of domain concepts and concepts from outside the domains and processes of interest. Further, during the specifying learning phase, we have extended our method with concepts, steps, and guidelines for grammatical analysis, for formalization of constraints, and for the specification of processes.

Survivorship wellness: a multidisciplinary group program for cancer survivors

PurposeNational mandates require cancer centers provide comprehensive survivorship care. We created an 8-session, group intervention, the Survivorship Wellness Group Program (SWGP), that covered 8 topics: nutrition, physical activity, stress, sleep/fatigue, sexuality/body image, emotional wellbeing/fear of cancer recurrence, spirituality/meaning, and health promotion/goal setting. This study examined the acceptability and preliminary outcomes of SWGP.MethodsWe evaluated SWGP using questionnaire data collected at program entry and 15-week follow-up. Questionnaires assessed acceptability and impact on anxiety, depression, quality of life, and perceived knowledge of topics. Enrollees who consented to participate in research and completed the baseline and 15-week follow-up were included in the analysis (N = 53). We assessed acceptability and preliminary outcomes using paired-samples t-tests. Due to the COVID-19 pandemic, SWGP transitioned to telehealth partway through data collection. Post-hoc analyses compared outcomes by intervention delivery.ResultsParticipants completed an average of 7.44/8 classes. Participants reported a mean response of 3.42/4 regarding overall program satisfaction and 90.6% reported being “very likely” to recommend SWGP. SWGP was associated with decreases in anxiety and depression; increases in physical, emotional, functional, and overall quality of life; and increases in knowledge of all health behavior domains. No outcomes differed significantly between delivery in person versus telehealth.ConclusionsSWGP offers an acceptable and replicable model for cancer centers to meet national survivorship care guidelines.Implication for cancer survivorsSWGP provides a comprehensive service for cancer survivors post-treatment, and was associated with better quality of life, fewer mental health symptoms, and increased knowledge in multiple domains of wellness.

IPlus a User-Centered Methodology for Serious Games Design

Standard video games are applications whose development process often follows a traditional software methodology. Serious Games (SGs) are a tool with an immensely positive impact and great success. SGs enable learning and provide entertainment and self-empowerment, which motivates students. The development of an SG consists of complex processes requiring multi-disciplinary knowledge in multiple domains, including knowing the learning domain and adding the appropriate game mechanics to foster high intrinsic motivation and positive player experience that makes the players feel like they are having fun while learning. Otherwise, the game is viewed as boring and not as a fun and engaging activity. Nevertheless, despite their potential, the application of SGs in education has been limited in terms of pedagogy. Several authors assert that this lack is because SG standards and guidelines have not been developed. There is an imbalance between experts’ contributions to education and game design specialists for the SG development. Not all the SGs that have been developed have applied appropriate design methodologies that incorporate both the entertainment mechanics and the serious component. To ensure that an SG meets the user’s expectations, it must be designed using an appropriate method. This work aims to present iPlus, a methodology for designing SGs based on a participatory, flexible, and user-centered approach. Additionally, this paper analyses several case studies with the iPlus methodology.

Incremental learning based multi-domain adaptation for object detection

Cross-domain object detection uses knowledge from source domain tasks to enhance the object detection in target domain. It can reduce the workload of data annotations in the new domain and significantly improve the adaptation ability of the network. We consider a more realistic transfer scenario, that is, our target domain samples cannot be obtained at the same time so that learners cannot forget the old domain knowledge while learning the new domain knowledge, which is consistent with the basic assumption of incremental learning. However, the existing methods do not consider the sequential learning process of the multiple target domains which may cause performance degradation. To address this issue, we propose a novel multi-domain adaptation method for object detection based on incremental learning. Specifically, the incremental learning network saves the knowledge of multiple domains and makes the model to fuse the knowledge of different domains during the training effectively. To make it work better, the progressive training strategy is proposed to make the model gradually adapt to multiple domains. Moreover, we use a multi-level feature alignment module to ensure that domain alignment is realized on features at various levels. We perform experiments on two sets of 6 datasets, which demonstrate that our model can effectively solve the problem of domain knowledge forgetting in multi-target domain adaptation and significantly improve the detection accuracy in each domain.

Multi-Source Transfer Learning Based on Inductive Knowledge-Leveraged for Medical Datasets

Transfer learning changes the limitation of the same probability distribution among domains. There are many innovative ideas of those models which are fully used the information and the knowledge from different domains. Additional knowledge by transferring learning is beneficial to improve the learning ability in target tasks. However, most multiple source domain transfer learning algorithms are developed for the specified model. The existing transfer TGHRR algorithm is suitable to one source only. Given this problem, a new multiple source transfer learning algorithm integrated with the TGHRR and the inductive knowledge of multiple domains (MS-TGHRR in brevity) is proposed. Furthermore, MS-TGHRR algorithm has been evaluated by experiments on medical datasets for classification task. Extensive experiments demonstrate the classification accuracies trained by the newly designed MS-TGHRR algorithm over the existing multiple source transfer learning algorithms.

Discovering Cross-Disciplinary Concepts in Multidisciplinary Context Through Collaborative Framework

When domain experts try to find a business solution, ambiguous terms arise within the context of discussion in a multidisciplinary research group. Different meanings and relationships with various concepts cause ambiguous semantics. This research aims to address complex business problems in a collaborative research group. This approach presents a collaborative framework based on network text analysis for detecting cross-disciplinary concepts in a multidisciplinary context. The framework recognizes ambiguous concepts (common terms presented in multiple domain knowledge), and these terms are visualized as a network. A case study of sustainable development demonstrates the identification of a set of cross-disciplinary concepts and their relationships across different domains. The main contributions are providing a framework to detect essential concepts that contain the cross-disciplinary concepts and recognize the understanding of multidisciplinary knowledge in the discussion context.

A topic-sensitive trust evaluation approach for users in online communities

In order to facilitate human decision making, trust evaluation has received widespread attention in many fields, especially for online services. Most of the existing methods consider trust in a person as a value which does not vary across different scenarios without any attention to the distinction of domains or communities where trust is derived. However, the notion of context is a significant and indispensable factor for trust evaluation in practice. Due to the lack of the consideration of context, traditional methods cannot resolve the issue that arises when a highly trustworthy person in one domain is likely to dominate the results of trust assessment in others where the person is in fact less authoritative. To solve this problem, in this paper, we develop a general approach to accomplish topic-sensitive trust evaluation by considering the context of trust. We first propose a general framework which presents the well-organized architecture of topic-sensitive trust evaluation in online communities. Then, a user-topic model is proposed to automatically extract topic data from user-generated content based on the Labeled Latent Dirichlet Allocation (LLDA) model. To compare the topic differences between users, we design a topic coverage function for revealing their trust relationships in diverse topics. Moreover, we employ two traditional methods and extend them to accomplish trust prediction for people with multiple domain knowledge. Experiments based on a real-world dataset show that extended topic-sensitive approaches are more adaptive and accurate than those topic-free trust evaluation approaches, especially when the trust application scenario features multiple topics.

From Help-Seekers to Influential Users: A Systematic Review of Participation Styles in Online Health Communities.

BackgroundUnderstanding how people participate in and contribute to online health communities (OHCs) is useful knowledge in multiple domains. It is helpful for community managers in developing strategies for building community, for organizations in disseminating information about health interventions, and for researchers in understanding the social dynamics of peer support.ObjectiveWe sought to determine if any patterns were apparent in the nature of user participation across online health communities.MethodsThe current study involved a systematic review of all studies that have investigated the nature of participation in an online health community and have provided a quantifiable method for categorizing a person based on their participation style. A systematic search yielded 20 papers.ResultsParticipatory styles were classified as either multidimensional (based on multiple metrics) or unidimensional (based on one metric). With respect to the multidimensional category, a total of 41 different participation styles were identified ranging from Influential Users who were leaders on the board to Topic-Focused Responders who focused on a specific topic and tended to respond to rather than initiate posts. However, there was little overlap in participation styles identified both across OHCs for different health conditions and within OHCs for specific health conditions. Five of the 41 styles emerged in more than one study (Hubs, Authorities, Facilitators, Prime Givers, and Discussants), but the remainder were reported in only one study. The focus of the unidimensional studies was on level of engagement and particularly on high-engaged users. Eight different metrics were used to evaluate level of engagement with the greatest focus on frequency of posts.ConclusionsWith the exception of high-engaged users based on high post frequency, the current review found little evidence for consistent participatory styles across different health communities. However, this area of research is in its infancy, with most of the studies included in the review being published in the last 2 years. Nevertheless, the review delivers a nomenclature for OHC participation styles and metrics and discusses important methodological issues that will provide a basis for future comparative research in the area. Further studies are required to systematically investigate a range of participatory styles, to investigate their association with different types of online health communities and to determine the contribution of different participatory styles within and across online health communities.

Expert Finding System using Latent Effort Ranking in Academic Social Networks

The dynamic nature of social network and the influence it has on the provision of immediate solutions to a simple task made their usage prominent and dependable. Whether it is a task of getting a solution to a trivial problem or buying a gadget online or any other task that involves collaborative effort, interacting with people across the globe, the immediate elucidation that comes into anyone's mind is the social network. Question Answer systems, Feedback systems, Recommender systems, Reviewer Systems are some of the frequently needed applications that are used by people for taking a decision on performing a day to day task. Experts are needed to maintain such systems which will be helpful for the overall development of the web communities. Finding an expert who can do justice for a question involving multiple domain knowledge is a difficult task. This paper deal with an expert finding approach that involves extraction of expertise that is hidden in the profile documents and publications of a researcher who is a member of academic social network. Keywords extracted from an expert's profile are correlated against index terms of the domain of expertise and the experts are ranked in the respective domains. This approach emphasizes on text mining to retrieve prominent keywords from publications of a researcher to identify his expertise and visualizes the result after statistical analysis.

プロブレムオントロジーに基づく医療行為の意図の表明支援

PROBLEM is a key concept used in sharing intentions of medical services among health professionals belonging to different domains. PROBLEMs represent the necessity for medical services from each domain. PROBLEMs serve as a reference point for unifying multiple domain knowledge, when health professional team designs services. However, there is confusion on sharing intentions of the medical services using PROBLEMs. The confusion arose from the ambiguity of the role which the PROBLEMs are playing on intention expressions. Hence, the authors developed a media equipped with a function which specifies roles of PROBLEMs, and tried to control of this confusion. The result of the trial use suggests that the media suppresses the confusion and prompts users to externalize intentions of medical services.

An ontology-based knowledge management framework for a distributed water information system

With the increasing complexity of hydrologic problems, data collection and data analysis are often carried out in distributed heterogeneous systems. Therefore it is critical for users to determine the origin of data and its trustworthiness. Provenance describes the information life cycle of data products. It has been recognised as one of the most promising methods to improve data transparency. However, due to the complexity of the information life cycle involved, it is a challenge to query the provenance information which may be generated by distributed systems, with different vocabularies and conventions, and may involve knowledge of multiple domains. In this paper, we present a semantic knowledge management framework that tracks and integrates provenance information across distributed heterogeneous systems. It is underpinned by the Integrated Knowledge model that describes the domain knowledge and the provenance information involved in the information life cycle of a particular data product. We evaluate the proposed framework in the context of two real-world water information systems.

A composite object model of an urban water network based on multidisciplinary knowledge bases. I. Composite object modeling

In order to model an urban water network, modelers have to cope with the interdisciplinary nature of the water network. To deal with the interdisciplinary nature, we have developed composite object modeling, a framework for constructing models on the basis of multiple domain knowledge bases. This modeling framework utilizes not only multiple domain knowledge, but also interface knowledge that bridges or interprets across disciplinary gaps among domains. A programming method combining object-oriented programming and logic programming is used to implement models on the framework. This method enables modelers to separate descriptions of the static structure of the models and those of the simulation procedures. By using the framework, we have also developed a knowledge system RARCOM (risk assessment and risk communication support system), which aims to support risk assessment and risk communication about the carcinogenicity of drinking water in an urban water network.

A multiple-domain knowledge based system: Coordination and integration of agronomic and economic knowledge bases with databases and bio-economic models

Abstract An integrated decision support system for pasture improvement was developed to assist consultants to help beef producers in the coastal subtropics decide on a pasture improvement strategy. The components of the system, including knowledge sources from two different domains (agronomy and economics), databases and static and dynamic bio-economic spreadsheet models, are described. The integration of these components was achieved through the use of a blackboard architecture. The system's interaction with users is illustrated by working through a case study. Another case study is used to discuss issues relating to the validation of multiple-domain decision support systems in agriculture.