Expert System Model Research Articles

A Proposed Model Expert System for Disease Diagnosis in Children to Make Decisions in First Aid

Children have a weaker immune system than adults. They are susceptible to disease. Therefore, this study proposes an expert system model of disease diagnosis in children. We develop expert strategies to meet the needs of alternative diagnostic tools in making decisions and first aid for children suffering from illness. The development of an expert system model for diagnosing children's diseases using forwarding chaining based on If-Then as an inference engine. We chose the forward chaining method because it has a framework for thinking like a doctor's when diagnosing and concluding the disease. We made testing to model by doctors with 35 patients. The test results show that the expert systems model of disease diagnosis in children in this research has to be used as an alternative or comparison diagnostic tool with an accuracy rate of 79%.

DSM-V Modelling as an Expert System Pilot in Classification of Insomnia Tendency Based on Time Range

Insomnia is a form of sleep disorder. This study develops an expert system model that can help determine the tendency of insomnia based on the Diagnostic and Statistical Manual for Mental Disorders (DSM-V) guidelines. The forward chaining design method was used in this study because it is bottom-up by collecting facts from patients and concluded based on the DSM-V guidelines. The forward chaining method was chosen to test the hypothesis of the classification of insomnia tendencies. The factual information then acts as a knowledge base fed into computer programs that can generate system rules. In addition, tacit knowledge is used, as evidenced by O'Leary validation, to strengthen the validation of psychologist practitioners. There are three validation criteria: the accuracy of the knowledge base, completeness of the knowledge base, and condition-decision matches. The parameters used are based on complaints, dysfunction, time range, and other factors. The results of modeling and analysis of the rule system using the forward chaining method classify insomnia tendencies into three types based on time range: episodic, persistent, and recurring. The validation results carried out by practicing psychologists based on the analysis showed that the three rule systems were following the DSM-V guidelines and practical experience.

The Design Model of English Graded Teaching Assistant Expert System Based on Improved B/S Three-Tier Structure System

With the continuous development of Internet technology, the modern education system is becoming increasingly intelligent. The integration of information technology with the traditional curriculum can unlock the potential of modern education, which is essentially a change from the traditional teacher-centered classroom teaching structure. The disadvantage of the traditional education system is that it is difficult to teach students according to their abilities because of the differences in English language standards caused by differences in their knowledge and learning potential. Although tiered teaching is one way of addressing this difference, it can have a negative impact in some ways, and implicit tiered teaching is a great solution to this problem. In addition, artificial intelligence is a science that studies the laws of human intelligence. The application of artificial intelligence technology to develop an expert system for teaching English can effectively solve the issues of insufficient teachers and a single teaching mode in the current education system. This will enable the sharing of teaching resources to effectively improve teaching quality, reduce the burden on teachers, and strengthen student’ ability to learn on their own. This paper introduces the principles of artificial intelligence and the B/S three-tier structure system, analyses the current problems faced by English teaching, and proposes a solution for the implementation of a university English teaching aid system based on the B/S three-tier structure system. This system is based on the teacher English language teaching assistant expert system module, which uses a combination of generative and framework-based knowledge representations of the domain knowledge of English language experts to better enable teachers to understand their students’ mastery of knowledge points. The results of this research show that the system can be used to help teachers understand students’ learning through reasoning and analysis, and to improve students’ interest, learning strategies and learning abilities in English.

Fuzzy logic expert system with multi-objective optimization of carbon nanotube infused nanopaint surface characteristics analysis using IRB 1410 robot

In this investigative study, the Taguchi design (L9) of experiments with TOPSIS multi-objective optimization is used for the robot nanopainting. The optimization objective is to maximize the film glue individually and minimize the surface blemishes and film thickness by varying the IRB 1410 robot painting parameters, such as the robot speed, pressure, and distance. The multi-wall carbon nanotube is infused into the paint materials using the ultrasonication process, and the mechanical properties are analysed with a comparison to regular paint materials. The virtual robot cycle time analysis is carried out for three different robot pathway patterns, namely linear, zig-zag, and circular. It is then compared with real-time experimental robot nanopaint on cold rolled close annealing steel materials. The obtained results show that surface roughness and thickness are minimized by 54% and 33.4%, respectively, using the robot nano spray coating when compared with regular spray coating. The film adhesive of robot Nano spray coating is maximized by 2.8% more than normal spray coating. A fuzzy logic expert system model is used to evaluate the surface finish characteristics of the nanopaint surface with low prediction error and with 89.2% confidence level. Heat transfer analysis of the robot nano-coated substrate is compared with that of the robot normal-coated substrate.

Fuzzy expert systems for prediction of ICU admission in patients with COVID-19

The pandemic COVID-19 disease has had a dramatic impact on almost all countries around the world so that many hospitals have been overwhelmed with COVID-19 cases. As medical resources are limited, deciding on the proper allocation of these resources is a very crucial issue. Besides, uncertainty is a major factor that can affect decisions, especially in medical fields. To cope with these issues, we use fuzzy logic (FL) as one of the most suitable methods in modeling systems with high uncertainty and complexity. We intend to make use of the advantages of FL in decisions on cases that need to treat in ICU. In this study, an interval type-2 fuzzy expert system is proposed for the prediction of ICU admission in COVID-19 patients. For this prediction task, we also developed an adaptive neuro-fuzzy inference system (ANFIS). Finally, the results of these fuzzy systems are compared to some well-known classification methods such as Naive Bayes (NB), Case-Based Reasoning (CBR), Decision Tree (DT), and K Nearest Neighbor (KNN). The results show that the type-2 fuzzy expert system and ANFIS models perform competitively in terms of accuracy and F-measure compared to the other diagnosis systems.

The Rationale and Approach of the Legal Expert System Construction

Based on the interpretation on the evolution of theoretical basis of the construction of legal expert system as the clue, this article mainly analyzes and points out the dilemma faced by symbolism and connectionism as the algorithm on construction of legal expert system. Through defeasible reasoning, a contextual-based legal expert system could be realized. The legal expert system should be constructed with the judge as the center character, in which the personality of the judge could not necessarily be eliminated. As a result, the article provides a model of a legal expert system based on dual processing, including factual argumentation and legal argumentation.

A method to enhance the flexibility of collaborative human-robot workspaces through an extended safety perspective

For the adaptation of a process in a flexible production environment, robotic systems are basically a suitable option to ensure the required agility. In case of a human-robot work system, however, the flexibility of a robot is currently severely limited due to safety reasons; therefore, new tasks for the robot can only be implemented with significant additional effort. In this paper we present a method for the flexibilization of collaborative robot applications within the constraints of robot safety as defined by ISO/TS 15066. We will show how an in-depth analysis of foreseeable tasks during the start-up phase of a collaborative human-robot work system provides the basis for both, a flexible as well as safe use. An expert system and suitable mathematical models ensure these modifications of tasks and changes in the workplace can be automatically evaluated in terms of safety. As a result, even non-safety-experts can make modifications and operate the system without imposing additional risks.

Artificial intelligence-powered expert system model for identifying fall armyworm infestation in maize (Zea mays L.)

Maize (Zea mays L) is one of the most saleable cereal crops grown worldwide and a dominant staple food in many developing countries. The severe outbreak of fall armyworm in maize causes massive yield loss. Modern technologies, including smartphones, can assist in detecting recognising the fall armyworm infestation in maize. The objective of this study was to develop an automated Artificial Intelligence Powered Expert System (AIPES) for identifying fall armyworm infestation in maize. In addition, it put forward a deep learning-based model that is trained on photographs of healthy and fall armyworm infested leaves, cobs and tassels from a dataset and furnished an application that will be detecting maize fall armyworm infestation using Convolutional Neural Network (CNN) architecture and Mobile Net V 2 framework model. The study developed an Artificial Intelligence (AI) based maize fall armyworm infestation detection system using a DCNN (Deep Convolutional Neural Network) to support maize cultivating farmers. The model executed the objective by accurately identifying the fall armyworm infested maize plant and also classified them vis-c-vis the healthier crop. The deep learning models were trained to detect and recognise fall armyworm infection using more than 11000 images of fall armyworm infested leaves, cobs, and tassels. The created application (AIPES for identifying fall armyworm infestation in maize) using CNN detected and recognised the fall armyworm infestation in maize with a 100 per cent training accuracy rate and 87 per cent validation accuracy. So, the detection of maize fall armyworm and the treatment of fall armyworm-infested maize could lead to a higher maize crop yield.

RSM based expert system development for cutting force prediction during machining of Ti–6Al–4V under minimum quantity lubrication

In recent days the manufacturing process have become more precise and cost efficient due to advancement in the field of computer technology. Information technology has been integrated with manufacturing practice and has resulted in time reduction from concept of a product to marketing of the product. Cutting force generated is the main manufacturing issue raised among industries as it clearly affects quality and cost of the final product. Hence using extensive literature and data base knowledge optimum cutting parameters are selected. Therefore, this paper focuses on a response surface methodology (RSM) based expert system that has been developed using JAVA programming with the help of response surface second order model to automatically generate values of cutting force during machining of Ti–6Al–4V alloy under minimum quantity lubrication (MQL) for different process input parameters. From RSM it has been observed that calculated value of F (20.36) was greater than the F-table value (3.02) and hence the model developed can be effectively used for machining of Ti–6Al–4V alloy. Further the developed RSM based expert system model can be successfully used to predict the force generated during cutting process while machining Ti–6Al–4V alloy under MQL conditions.

(Invited) Exploring the Potential of Machine Learning and AI for ALD Process Optimization

Over the past years there has been a surge of interest on the application of machine learning, and more broadly, artificial intelligence, to scientific applications. This interest has been spurred by an advancement of our computational capabilities coupled with the development of a large set of methods and machine learning frameworks to carry out complex analysis of large datasets in ways that were not possible in the past. In this presentation, I will explore the application of machine learning techniques to the area of atomic layer deposition, and in particular to process optimization. While most ALD processes are highly reproducible, the optimization of both new and existing ALD processes in novel reactors is often time consuming and labor intensive. This motivated us to search for ways of accelerating this task using both ex-situ and in-situ experimental data that is readily available in many laboratories.First, I will focus on the development of surrogate models that can predict the saturation dose times based on partially saturated growth profiles, a type of data that is part of the routine characterization of ALD processes in many labs. In order to develop this surrogate model, we first created a dataset using simulated data providing growth profiles across one of our ALD reactors for a wide range of possible experimental conditions, including precursor pressure, surface reactivity, growth per cycle, flow conditions, and dose times. We then used this dataset to train deep neural networks to predict the dose time that would lead to complete saturation. By tailoring the complexity of the model, the number of growth profiles per condition, and the structure/depth of the neural network we can explore the minimum number experiments and data points required to accurately predict saturation for an unknown ALD process. Our results show that for the optimization of a single precursor, neural networks can accurately predict saturation times using just a single growth profile and its corresponding dose time as an input.Then, I will focus on the use of in-situ techniques for the automatic optimization of ALD processes without a human in the loop. The advantage of using in-situ techniques is that they can provide a much faster response, reducing optimization times by more than two orders of magnitude. Using a realistic model of a single quartz crystal microbalance, we have developed both expert systems and Bayesian optimization-based models that are capable of exploring the process conditions and converge on optimal conditions leading to saturation. We then implemented the expert system model in one of our experimental reactors, experimentally validating this approach. This methodology can be expanded to other in-situ techniques, such as spectroscopic ellipsometry, or more sophisticated arrangement including more than one probe to obtain spatially dependent data.Finally, I will briefly discuss other potential applications of machine learning, for instance for process optimization using spatial ALD configurations, or to predict the self-limited nature of novel processes.This material is based upon work supported by Laboratory Directed Research and Development (LDRD) funding from Argonne National Laboratory, provided by the Director, Office of Science, of the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.

Rule-Based Expert System for Assist Physician to Diagnosis of Malaria: XPerMal

Nowadays, common diseases like malaria, typhoid and cholera become more dangerous problems for people living in this world. The objective is how it can avoid the queue of patients in hospital. In this article, the author has proposed a model of expert systems using the knowledge of physician and other health professionals. The rule based expert system XPerMal useful for patients infected with common diseases and this system will give an answer as similar to a doctor or medical expert and also this system is very useful in rural areas where we have young medical experts or have no medical expert. The reasoning strategy is a key element in many medical tasks. It is well known that developing countries face a shortage of medical expertise in the medical sciences. Patients also find a huge queue in hospitals. Because of this, they are unable to provide good medical services to their inhabitants. The knowledge is acquired from literature review and human experts in the specific field and is used as a basis for analysis, diagnosis and decision-making. Knowledge is represented by an integrated formalism that combines rules and facts.

An Ensemble Learning Algorithm for Machinery Fault Diagnosis Based on Convolutional Neural Network and Gradient Boosting Decision Tree

In the massive mechanical fault data, the value density of fault information is low, and the data quality is uneven. What’s more, the multi-source signals collected by different sampling methods are different. At present, the expert system or shallow neural network model with weak self-learning ability cannot meet the requirements. Therefore, aiming at the characteristics of coupling, uncertainty and concurrency of mechanical faults, this paper constructs two kinds of 2D-CNN fault feature data sets, and uses Convolutional Neural Network (CNN) with strong self-learning ability to build two kinds of fault diagnosis models: CNN-Z and CNN-F. With CNN-Z and CNN-F models as base learners, this paper utilizes the ensemble algorithm Gradient Boosting Decision Tree (GBDT) to combine multiple bases. Compared with the results of the single base learner, the outcomes have higher accuracy and lower generalization error. Through the analysis and comparison of the performance indicators of the algorithm, this paper concludes that the diagnosis error of the fault diagnosis algorithm based on CNN and GBDT is the lowest with 1.79%, and the effectiveness, reliability and accuracy of the proposed algorithm in mining hidden fault state information are verified.

A Survey of Belief Rule-Base Expert System

The belief rule-base (BRB) model is a new intelligent expert system with the characteristics of both expert system and data-driven model. In BRB there are many if-then rules which use belief degrees to express various types of uncertain information, including fuzziness, randomness, and ignorance. As a semi-quantitative modeling tool for complex systems, BRB has the superiorities of dealing both numerical quantitative data and linguistic qualitative knowledge that are derived from heterogeneous sources. Moreover, it is also a white box approach which can provide direct access and transparency to decision makers and stakeholders. Currently, BRB has been widely applied in many fields, such as decision making, reliability evaluation, network security situation awareness, fault diagnosis, and so on. To fully demonstrate the progress of BRB, the original BRB, and some evolution forms are introduced in this article.

Fuzzy-Based Application Model and Profile Matching for Recommendation Suitability of Type 2 Diabetic

Diabetes Mellitus (DM) is a metabolic disease characterized by hyperglycemia due to insulin secretion abnormalities and a global health threat. DM has several types, namely type 1, 2, gestational, and other types. Type 2 diabetes patients have the largest number in the world. DM therapy can be done in 2 (two) ways: improving lifestyle and administering drugs. The problems and risks in recommending drugs are essential in the patient's healing process because they are likely to take medicine for life. Approximately 260,000 patients with type 2 diabetes experienced medication errors in 2017. The doctor's mistake in recommending drugs causes a long healing process and costs more. Recommending drugs requires pharmacological knowledge, and not all hospitals have pharmacologists. Several researchers have researched recommendations for antidiabetic drugs, but no studies have yet been found that discuss recommendations for combination antidiabetic drugs for type two to determine dosage and frequency. The number of medications used is 6 to 7, with many parameters 5 to 8. The latest endocrinology guidelines for 2020 state that in recommending antidiabetic drugs, not only 6 to 7 participants, but still need to maintain other aspects. Therefore, this study aims to build an expert system model with a new approach in recommending antidiabetic drugs with more complete parameters and recommend dosage and frequency. The model developed uses the Fuzzy Profile Matching method. Fuzzy is used to calculate the suitability between the patient's condition and the type of antidiabetic drug. Profile Matching is used to calculate the core factor and secondary factor to obtain each drug's total value. The dose was calculated using the FIS Tsukamoto for inputting low dosage, and high dosage calculated the weighted average value. Determination of frequency using the IF-Then function. Model evaluation is done by comparing recommendation data from doctors. The results of the evaluation of the model obtained an accuracy of 90%. This system will reduce medical personnel errors in recommending antidiabetic drugs that can positively impact patients' time, the healing process, and costs. This study provides knowledge that antidiabetes drugs' determination requires many parameters, while other studies used only 4 to 8. This study also provides an overview of the dosages of drugs that drug companies can produce. Usually, the company only makes low and high dosage. This study shows that creating multiple drug dosage is more efficient for patients.

Research on Multi-Target Network Security Assessment with Attack Graph Expert System Model

With the rapid development of the Internet, network attacks often occur, and network security is widely concerned. Searching for practical security risk assessment methods is a research hotspot in the field of network security. Network attack graph model is an active detection technology for the attack path. From the perspective of the attacker, it simulated the whole network attack scenario and then presented the dependency among the vulnerabilities in the target network in the way of directed graph. It is an effective tool for analyzing network vulnerability. This paper describes in detail the common methods and tools of network security assessment and analyzes the construction of theoretical model of attack graph, the optimization technology of attack graph, and the research status of qualitative and quantitative analysis technology of attack graph in network security assessment. The attack graph generated in the face of large-scale network is too complex to find the key vulnerability nodes accurately and quickly. Optimizing the attack graph and solving the key attack set can help the security manager better understand the security state of the nodes in the network system, so as to strengthen the security defense ability and guarantee the security of the network system. For all kinds of loop phenomena of directed attribute attack graph, the general method of eliminating loop is given to get an acyclic attack graph. On the basis of acyclic attack graph, an optimization algorithm based on path complexity is proposed, which takes atomic attack distance and atomic weight into consideration, and on the basis of simplified attack graph, minimum-cost security reinforcement is carried out for the network environment. Based on the ant colony algorithm, the adaptive updating principle of changing pheromone and the local searching strategy of the adaptive genetic algorithm are proposed to improve the ant colony algorithm. The experimental results show that compared with the ant colony algorithm, the improved ant colony algorithm can speed up the process of solving the optimal solution. When the number of attack paths is large, the advantages of the improved ant colony algorithm in solving accuracy and late search speed are more obvious, and it is more suitable for large-scale networks.

Neuro-fuzzy algorithm for clustering multidimensional objects in conditions of incomplete data

The paper is considered development of fuzzy expert system model for identifying faults in complex systems using data mining methods based on searching for hidden patterns in databases. The use of neural network technologies makes it possible to detect nonlinear dependencies of input and output data, improve the quality of an objective assessment of the state of complex technical objects, which ultimately will reduce the number of emergency situations during operation. A method is proposed for identifying the optimal number of fuzzy clusters in the space of training examples and determining, on their basis, the parameters of the membership functions for the input variables and inference results. Considered a neuro-fuzzy algorithm for clustering multidimensional objects in conditions of incompleteness and fuzzy initial information.

Comparison of the performance of two grades of metformin hydrochloride elaboration by means of the SeDeM system, compressibility, compactability, and process capability indices

Quality by design, applied to the development of a pharmaceutical drug, demands scientific methodologies, representing a source of information that will allow for a complete understanding the production process and the materials used for its manufacturing. Although the SeDeM system is a tool that enables a rational development of a product, result does not assure that an assessed material or mixture will be successful in terms of compression, hence, further research will be necessary on these features. The objective of this study was to assess and compare two grades of metformin hydrochloride elaboration: crystalline and direct compression using PXRD, the SeDeM expert system, the Heckel and Ryshkewitch-Duckworth models, as well as process control tools such as control charts and process capability indices to characterize and predict the performance of the materials in a direct compression process. The assessment identified that in spite of dealing with two different technical grades of a material with specific critical quality attributes for each one, PXRD analysis showed we dealt with the same crystalline structure, while the SeDeM system profiles obtained have very close values, and the main differences in materials were observed when subjecting them to conditions that simulate a compaction process with the Ryshkewitch-Duckworth model, in which a 46-times higher mechanical resistance was observed in the direct compression material compared with the crystalline one. The statistical control analysis revealed that only the direct compression material could be used to elaborate tablets whose weight variation was always maintained within the specification and control limits.

Application of LSTM Networks for Water Demand Prediction in Optimal Pump Control

Every morning, water suppliers need to define their pump schedules for the next 24 h for drinking water production. Plans must be designed in such a way that drinking water is always available and the amount of unused drinking water pumped into the network is reduced. Therefore, operators must accurately estimate the next day’s water consumption profile. In real-life applications with standard consumption profiles, some expert system or vector autoregressive models are used. Still, in recent years, significant improvements for time series prediction have been achieved through special deep learning algorithms called long short-term memory (LSTM) networks. This paper investigates the applicability of LSTM models for water demand prediction and optimal pump control and compares LSTMs against other methods currently used by water suppliers. It is shown that LSTMs outperform other methods since they can easily integrate additional information like the day of the week or national holidays. Furthermore, the online- and transfer-learning capabilities of the LSTMs are investigated. It is shown that LSTMs only need a couple of days of training data to achieve reasonable results. As the focus of the paper is on the real-world application of LSTMs, data from two different water distribution plants are used for benchmarking. Finally, it is shown that the LSTMs significantly outperform the system currently in operation.

Modeling and simulation of a multi-parametric fuzzy expert system for variable rate nitrogen application

Nitrogen (N) excess due to mineral fertilization in conventional crop farming has a significant negative impact on the environment. Variable rate N application (VRNA) is a promising tool to increase N recovery rates in spatially heterogeneous fields. Real-time sensor systems for VRNA usually consider only the crop’s N status and their fertilization algorithms are abundantly deterministic. Due to their education and professional experience, farmers have a considerable knowledge base that should be used to describe the dynamic and non-deterministic interactions of multiple parameters for a locally adapted N fertilization. Fuzzy systems present an effective way to integrate expert knowledge into an automated multi-parametric control. This paper describes, how fuzzy logic can be used to fuse the plant-related information from a real-time sensor system with further parameters to create a multi-parametric system for VRNA. Using sets of input–output data acquired with a Yara N-Sensor ALS2 system, an adaptive, fuzzy logic-based model of its agronomic algorithms was identified, optimized and validated. The results indicated high accordance with the N-Sensor algorithms and good automated adaptability to different calibrations with values of the Pearson correlation coefficient higher than 0.99 and a maximum percentage root mean square error of 0.14%. In a case study, the model was combined with the apparent soil electrical conductivity (ECa) as an indicator for spatially varying soil productivity, as well as a case distinction for different weather conditions. Simulations with historic ECa data and N-Sensor recordings have shown the high flexibility of the multi-parametric fuzzy expert system. With the presented method, specific deficiencies of one-parametric approaches can be moderated and the application can be adapted to the prevailing conditions in a straightforward manner. Also, the target orientation could be influenced based on the specific preferences of the expert.

Modeling Academic Research Collaborator Selection Using an Integrated Model

Expert finding systems try to alleviate the information overload problem and recommend experts who can satisfy users' needs. They support researchers to find research collaborators automatically. The main challenge of current expert finding systems is that they retrieve experts based on the content of their documents but ignore the human interaction perspective. The human interaction perspective comprises the factors that influence collaborator selection decisions in real life. This study aimed to develop a collaborator selection model for expert finding systems in research universities. This model includes human capital, social capital, and cultural capital factors that influence collaborator selection. The researchers integrated the Scientific and Technical Human Capital (STHC) model and Social Capital Theory to determine these factors. The authors conducted a survey comprising 349 researchers from Malaysian research universities to validate the research hypotheses. A partial least squares structural equation model (PLS-SEM) was employed to analyze all the survey data. The empirical results revealed that the significant factors that influence collaborator selection in the research universities context were cognitive accessibility, reliability, relevance, commitment, physical accessibility, cultural experiences, complementary skills, and research experience. Surprisingly, the results revealed that network ties, relational accessibility, and reputation were insignificant factors for collaborator selection. This study proposed a research model for collaborator selection in the research universities context and provided several recommendations for the policymakers and practitioners. The model will help to provide sufficient criteria to select academic research collaborator in universities and can be used by expert finding systems designers, researchers, collaborators, and universities.