Insufficient Domain Knowledge Research Articles

Clustering approximation via a fusion of multiple random samples

In big data clustering exploration, the situation is paradoxical because there is no prior or insufficient domain knowledge. Moreover, clustering a big dataset is a challenging task in distributed computing framework. To address this, we propose a new distributed clustering approximation framework for big data with quality guarantees. This innovative framework uses multiple disjoint random samples instead of a single random sample to compute an ensemble result as the estimation of the true result of the entire big dataset. To begin, we modeled a large dataset as a collection of random sample data blocks stored in a distributed file system. Henceafter, a subset of data blocks is randomly selected, and to generate the component clustering results, the serial clustering algorithm is executed in parallel on the distributed computing framework. In each selected random sample, the number of clusters and initial centroids is identified using a density peak-based I-niceDP clustering algorithm, and then the k-means sweep refines them. Since the random samples are disjoint and traditional consensus functions cannot be used, we propose two new methods, a graph similarity and a naturally inspired firefly-based algorithm, to integrate the component clustering results into the final ensemble result. The entire clustering process is displayed through systematic support, extensive measures of clusterability, and quality evaluation. The methods are verified in a series of experiments using synthetic and real-world datasets. Our comprehensive experimental results demonstrate that the proposed methods vividly (1) recognize the correct number of clusters by analyzing a subset of samples and (2) exhibit better scalability, efficiency, and clustering stability.

Detection of Diseases in Pandemic: A Predictive Approach Using Stack Ensembling on Multi-Modal Imaging Data

Deep Learning (DL) in Medical Imaging is an emerging technology for diagnosing various diseases, i.e., pneumonia, lung cancer, brain stroke, breast cancer, etc. In Machine Learning (ML) and traditional data mining approaches, feature extraction is performed before building a predictive model, which is a cumbersome task. In the case of complex data, there are a lot of challenges, such as insufficient domain knowledge while performing feature engineering. With the advancement in the application of Artificial Neural Networks (ANNs) and DL, ensemble learning is an essential foundation for developing an automated diagnostic system. Medical Imaging with different modalities is effective for the detailed analysis of various chronic diseases, in which the healthy and infected scans of multiple organs are compared and analyzed. In this study, the transfer learning approach is applied to train 15 state-of-the-art DL models on three datasets (X-ray, CT-scan and Ultrasound) for predicting diseases. The performance of these models is evaluated and compared. Furthermore, a two-level stack ensembling of fine-tuned DL models is proposed. The DL models having the best performances among the 15 will be used for stacking in the first layer. Support Vector Machine (SVM) is used in Level 2 as a meta-classifier to predict the result as one of the following: pandemic positive (1) or negative (0). The proposed architecture has achieved 98.3%, 98.2% and 99% accuracy for D1, D2 and D3, respectively, which outperforms the performance of existing research. These experimental results and findings can be considered helpful tools for pandemic screening on chest X-rays, CT scan images and ultrasound images of infected patients. This architecture aims to provide clinicians with more accurate results.

Teaching and Learning Domain Modeling through Collaboration Patterns: A Controlled Experiment

Domain models in software engineering—often represented as class diagrams—depict relevant classes in a given problem domain along with necessary relationships among those classes. These models are important because they establish links between the requirements of a given system under development and the subsequent phases of the systems development life cycle. Although the teaching of basic concepts related to domain modeling takes only about 1 or 2 hours, proper application of these concepts to a given problem situation is difficult for students studying software engineering. Due to their insufficient domain knowledge of the problem situation and modeling experience, they often produce domain models that may not adequately represent necessary elements as part of the domain models. Analysis patterns can help them by encoding expert knowledge and offering guidance in the modeling process. This article reports the findings from a controlled experiment conducted to study the effects of collaboration patterns on the domain modeling process by students. Specifically, the study investigated the differences in students’ perceptions of the ease of the domain modeling process and quality of models produced, perceived difficulties, and how collaboration patterns help address domain modeling difficulties and the quality of domain models produced. Findings from this experimental study involving students from a software engineering course indicate that although there is no significant difference in subjects’ perceptions between the control and treatment groups, the subjects from the treatment group produced better-quality domain models. Additionally, the qualitative analysis of the feedback collected from the subjects from the control and treatment groups reveals that that having knowledge of patterns is beneficial, as it addresses the difficulties in domain modeling.

Contingency response decision of network public opinion emergencies based on intuitionistic fuzzy entropy and preference information of decision makers

A multi-attribute group decision-making (MAGDM) method based on intuitionistic fuzzy preference information is proposed for the multi-attribute intuitionistic fuzzy group decision-making problem where the decision-makers weight and attribute weight are completely unknown and the decision-maker has preference information for the scheme. Firstly, an intuitionistic fuzzy interval judgment matrix is established to describe the original data of the key decision indicators for multiple network public opinion emergencies that erupt simultaneously. Secondly, the attribute weights are determined based on the improved intuitionistic fuzzy entropy construction method, and the expert weights are determined by using objective decision information, taking into account the intuitionistic fuzzy entropy of decision matrix. MAGDM can not only synthesize experts' professional experience in different aspects, but also avoid experts' decision-making errors caused by insufficient domain knowledge. It is necessary to continuously adjust the expert weight based on decision-making information to make the comprehensive decision-making information more accurate. Thirdly, a scheme preference model and an attribute weight optimization model are established to determine the ranking method of intuitionistic fuzzy interval values. Then, a modified distance measure of intuitionistic fuzzy sets (IFSs) is introduced to make the evaluation result more accurate and reasonable when it comes to solving the deviation between the evaluation value and ideal solution of each scheme. Finally, the effectiveness and practicability of the proposed decision-making method are verified by an example of emergency crisis severity, It assists decision makers in selecting and implementing the optimal emergency response plan in a timely and efficient manner, which improves the emergency treatment efficiency of network public opinion crisis, helps emergency departments to better deal with the network public opinion crisis, improves the ability of public opinion guidance and control, and provides a new method and idea for multi-attribute intuitionistic fuzzy group decision-making problem.

Extroversion Personality, Domain Knowledge, and the Creativity of New Product Development Engineers

Creativity refers to a person’s thinking of new and useful ideas at work. Drawing on the personality literature, this study proposes that employees with high extroversion personality will exchange job relevant information with colleagues to generate creativity. Drawing on the information exchange theory, this study further proposes that the effect of extroversion personality with creativity via information exchange will be stronger for workers with insufficient domain knowledge. Data collected from Taiwanese new product development engineers support our hypotheses. Previous research found that extroverted persons perform more creatively because they are more confident in their abilities. This study further demonstrates that, after controlling for self-efficacy, extroverted engineers can still think of creative ideas via exchanging information with colleagues. Although domain knowledge has been shown as crucial to creativity, few studies have explored how to stimulate creative ideas from workers with insufficient domain knowledge. This study demonstrates a useful substitute for domain knowledge, namely the extroversion personality, which may inspire creativity via exchanging information with colleagues.

From syntactic-semantic tagging to knowledge discovery in medical texts

In the GALEN project, the syntactic-semantic tagger MultiTALE is upgraded to extract knowledge from natural language surgical procedure expressions. In this paper, we describe the methodology applied and show that out of a randomly selected sample of such expressions coming from the procedure axis of Snomed International, 81% could be analysed correctly. The problems encountered fall in three different categories: unusual grammatical configurations within the Snomed terms, insufficient domain knowledge and different categorisation of concepts and semantic links in the domain and linguistic models used. It is concluded that the Multi-TALE system can be used to attach meaning to words that not have been encountered previously, but that an interface ontology mediating between domain models and linguistic models is needed to arrive at a higher level of independence from both particular languages and from particular domains.