Chantamit-O-Pass, Pattanapong
Loading...
Preferred name
Chantamit-O-Pass, Pattanapong
Alternative Name
Chantamit-O-Pas, Pattanapong
Main Affiliation
Email
pattanapong.ch@kmitl.ac.th
2 results
Now showing 1 - 2 of 2
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Long Short-Term Memory Recurrent Neural Network for Stroke Prediction(2018-01-01); Goyal, MadhuElectronic Healthcare Records (EHRs) describe the details about a patient’s physical and mental health, diagnosis, lab results, treatments or patient care plan and so forth. Currently, the International Classification of Diseases, 10<sup>th</sup> Revision or ICD-10 code is used for representing each patient record. The huge amount of information in these records provides insights about the diagnosis and prediction of various diseases. Various data mining techniques are used for the analysis of data deriving from these patient records. Recurrent Neural Network (RNN) is a powerful and widely used technique in machine learning and bioinformatics. This research aims at the investigation of RNN with Long Short-Term Memory (LSTM) hidden units. The empirical research is intended to evaluate the ability of LSTMs to recognize patterns in multi-label classification of cerebrovascular symptoms or stroke. First, we integrated ICD-10 code into health record, as well as other potential risk factors within EHRs into the pattern and model for prediction. Next, we modelled the effectiveness of LSTMs for prediction of stroke based on healthcare records. The results show several strong baselines that include accuracy, recall, and F1 measure score. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Knowledge Acquisition, Representation and Transfer in the Metamodel for Software Development Methodologies(2019-10-01)At present, software development life cycle has many standards involved. However, during the process of creating components in an application, there are many problems that originate in the knowledge gap between stakeholders and developers. The problem of developing is issues such as ambiguity, misunderstanding, miscommunications, different domain skills, and conflicting goals. Metamodel are needed because the methodology contents must be purpose-fit and optimally adjusted for the users/stakeholders/organisation. Metamodel, a model of software model includes endeavor domain, method domain, and metamodel domain. It has contents for user's needs, including organization, project, and product aspect. Three interconnected meta-elements are work products, work units, and producers, as well as relationship between them. ISO/IEC 24744 is the standard that includes metamodels.Application software are now applied knowledge management components aiming to reduce the gap between users/stakeholders and software developer team. The knowledge can be shared and transferred to others. This paper, therefore, focuses on applying knowledge management metamodel for software development process. We will extend the metamodel concept in ISO/IEC 24744 for knowledge acquisition, knowledge representation, and knowledge transfer to WorkUnits and WorkProducts, thereby increasing their explicitness in the software development process. Tacit knowledge will be focused and propose knowledge metamodel will be shown.