Neural Network-based Expert System Research Articles

IoMT-Based Automated Diagnosis of Autoimmune Diseases Using MultiStage Classification Scheme for Sustainable Smart Cities

The resolution of complex medical diagnoses using pattern recognition requires an artificial neural network-based expert system to automate autoimmune disease diagnosis in blood samples. This process is done using image-based computer-aided diagnosis (CAD) to reduce errors in the diagnosis process. This paper describes a Multistage Classification Scheme (MSCS), which uses antinuclear antibody (ANA) tests to identify and classify the existence of autoantibodies in the blood serum that bind to antigens found in the nuclei of mammalian cells. The MSCS classified HEp-2 cells into three stages by using Binary Tree (BT), Artificial Neural Network (ANN), and Support Vector Machine (SVM) as basic blocks. The Indirect Immunofluorescence (IIF) technique is used in the ANA test with Human Epithelial type-2 (HEp-2) cells as substrates. The efficiency of the proposed methodology is assessed using the dataset of ICPR 2016. The intermediate cells (IMC) and positive cells (PC) were separated in Stage 1 prior to preprocessing based on their total strength, and special preprocessing is applied to intermediate cells for improved output, and positive cells are subjected to mild preprocessing. The mean class accuracy (MCA) was 84.9% for intermediate cells and 95.8% for positive cells, although the carefully picked 24 features and SVM classifier were applied. ANN showed better performance by adjusting the weights using the SCGBP algorithm. So, the MCA is 88.4% and 97.1% for intermediate and positive cells, respectively. BT had an MCA of 95.3% for intermediate and 98.6% for positive. In Stage 2, the meta learners BT2, ANN2, and SVM2 were trained for an augmented feature set (24 + 3 results from base learners). Therefore, the performance of BT2, ANN2, and SV M2 was increased by 1.8%, 4.5%, and 4.1% as compared to Stage 1. In Stage 3, the final prediction was performed by majority voting among the results of the three meta learners to achieve 99.1% MCA. The proposed algorithm can be embedded into a CAD framework built for the ANA examination. The proposed model will improve operational efficiency, decrease medical expenses, expand accessibility to healthcare, and improve patient safety in the sector, enabling enterprises to lower unplanned downtime, develop new products or services, increase operational effectiveness, and enhance risk management.

Enabling laser transmission welding of additively manufactured thermoplastic parts using an expert system based on neural networks

In order to use laser transmission welding (LTW) for additively manufactured parts such as prototypes, small series, or one-off products, an enhanced process knowledge is needed to overcome the difficulties in the part composition resulting from the additive manufacturing process itself. In comparison to an injection molding process for thermoplastic parts, the additive manufacturing process fused deposition modeling leads to an inhomogeneous structure with trapped air inside the volume. In this paper, a neural network-based expert system is presented that provides the user with process knowledge in order to improve the weld seam quality of laser welded additively manufactured parts. Both additive manufacturing and LTW process are assisted by the expert system. First, the designed expert system supports the user in setting up the additive manufacturing process to increase the transmissivity. During welding, the additive manufacturing and LTW process parameters are used to predict the weld seam strength. To create the database for the expert system, specimens of transparent and black polylactide are additively manufactured. In order to change the transmissivity at an emission wavelength of 940 nm of the diode laser used, the manufacturing parameters for the transparent parts are varied. The transmissivity of the parts is measured with a spectroscope. The transparent samples are welded to the black samples with laser powers between 8 and 14 W in the overlap configuration and shear tensile tests are performed. In this work, the predictions of the transmissivity and the shear tensile force are demonstrated with an accuracy of more than 88.1% of the neural networks used for the expert system.

Expert system-supported optimization of laser welding of additively manufactured thermoplastic components

Laser transmission welding (LTW) is a known technique to join conventionally produced thermoplastic parts, e.g. injected molded parts. When using LTW for additively manufactured parts (usually prototypes, small series), this technique has to be evolved to overcome the difficulties in the part composition resulted in the additive manufacturing process itself.In this paper, a method is presented to enhance the weld seam quality of laser welded additively manufactured parts assisted by a neural network-based expert system. To validate the expert system, specimens are additively manufactured from polylactide. The parameters of the additive manufacturing process, the transmissivity, and the LTW process parameters are used to predict the shear tensile force with the neural network. The transparent samples are welded to black absorbent samples in overlap configuration and shear tensile tests are performed. In this work, the prediction of the shear tensile force with an accuracy of 88.1% of the neuronal network based expert system is demonstrated.

Design of Neural Network Based Expert System for Automated Lime Kiln System

Design of Neural Network Based Expert System for Automated Lime Kiln System

Neural network-based expert system for modeling of tube spinning process

The present paper deals with the development of neural network (NN)-based expert system for modeling of 2024 aluminum tube spinning process. Tube spinning is a highly nonlinear thermo-mechanical process for producing large-diameter thin-walled shapes. It is interesting to note that the performance of the process depends on various process parameters, such as wall thickness, percentage of thickness reduction, feed rate, mandrel rotational speed, solution treatment time and aging time. Therefore, an NN-based expert system is necessary for modeling the tube spinning process. The input layer of NN consists of six neurons corresponding to the inputs of the tube spinning process. Moreover, the output layer consists of four neurons that represent four responses, namely change in diameter, change in thickness, inner and outer surface roughness. It is to be noted that the performance of NN depends on various factors, such as number of neurons in the hidden layer, coefficients of transfer functions and connecting weights, etc. In the present paper, three algorithms, such as back-propagation, genetic and artificial bee colony algorithms, are used for optimizing the said variables of NN. Further, the developed approaches are tested for their accuracy in prediction with the help of some test cases and found to model the tube spinning process effectively.

Computing the tensile behaviour of tailor welded blanks made of dual-phase steel by neural network-based expert system

This work primarily aims to develop an expert system based on the artificial neural network (ANN) to predict the tensile behaviour of tailor welded blanks (TWBs) made of dual-phase (DP) 590 steel. The work also aims to compare the predictions by ANN models with empirical models and the size of the training data set of the prediction accuracy of these models. The strain hardening exponent ‘n’ and strength coefficient ‘K’ are predicted. The results obtained from expert system and empirical models are validated by comparing them with the results obtained from finite element simulations and experiments. It is observed that expert system/ANN predictions based on the full factorial design of experiments (DOE) is better than the ANN predictions based on the orthogonal array and predictions based on the empirical models. With the reduced orthogonal training data, ANN model-based predictions are more accurate than the empirical models in most of the test cases taken outside the training range. Inverse models for predicting the TWB parameter combination for good tensile characteristics are also developed and show promising results. The ANN/expert system developed through full factorial DOE is comparable with the experimental results. The ANN/expert system developed through orthogonal DOE is not comparable with the experimental results.

Neural Network Based Expert System for Steel Bar Pipeline Fault Diagnosis

In modern continuing steel bars production pipeline, there are various equipments and automatic controls combining with mechanism, electricity, hydraulic pressure and aerodynamic. This field of fault detection and diagnosis deals with design of computer-based automated system that can assist plant operators. The neural network based expert system have advantages of parallel distributed processing, high robust, fault tolerance, adaptive and self-organization. Applying neural network based expert system for the condition detection and fault diagnosis of steel bars pipeline can reduce the economic loss caused by system downtime.

Application of neural network-based expert system in drilling fault diagnosis

Combining the practical conditions and the drilling variables, the drilling fault diagnosis system utilizing the Artificial Neural Network (ANN) to obtain the knowledge and the expert system to reason and diagnose the faults was designed. The knowledge was acquired through constantly learning and training the ANN with varieties of the intricate drilling problems. The definite diagnosis results can be drawn by the reverse reasoning of the expert system. According to the results the drilling parameters will be adjusted automotively. The drilling fault diagnosis system has proved to be practical and effective by the living examples.

Assessing Customer Satisfaction in a V-commerce Environment

This article describes the design and implementation of a non-intrusive method of assessing customer satisfaction in a voice-enabled electronic commerce environment. After recording a customer's speech voice during his or her interaction with a voice-enabled Web system (VWS), a subsequent questionnaire survey was immediately administered to identify the satisfaction level of the customer. Afterward, a collection of recorded customer voice files and the corresponding values of customer satisfaction were used to construct an artificial neural network-based expert system, the satisfaction level assessment system (SLAS), which was thereafter integrated into VWS for automatically detecting the satisfaction level of VWS users. Experiments were performed to test the feasibility and applicability of the proposed method, and good preliminary results were derived. Instead of using the conventional questionnaire-based approach, SLAS is non-intrusive because it does not require users to fill out any questionnaire. The proposed method can be used by various voice-based business applications, such as call centers and customer relationship management, to achieve the business objective of improving customer satisfaction, enforcing customer loyalty, increasing re-purchase rate, and enhancing enterprise's benefits. The proposed SLAS (including method and system) that was filed for patent application was recently approved by the Taiwan Intellectual Property Office under Patent No. I268478.

An artificial neural network-based expert system for the appraisal of two-car crash accidents

This paper employs artificial neural network (ANN) to develop an accident appraisal expert system. Two ANN models – party-based and case-based – with different hidden neurons are trained and validated by k-fold ( k = 3) cross validation method. A total of 537 two-car crash accidents (1074 parties involved) are randomly and equally divided into three subsets. For the comparison, a discrimination analysis (DA) model is also calibrated. The results show that the ANN model can achieve a high correctness rate of 85.72% in training and 77.91% in validation and a low Schwarz's Bayesian information criterion (SBC) of −0.82 in training and 0.13 in validation, which indicates that the ANN model is suitable for accident appraisal. Furthermore, in order to measure the importance of each explanatory variable, a general influence (GI) index is computed based on the trained weights of ANN. It is found that the most influential variable is right-of-way, followed by location and alcoholic use. This finding concurs with the prior knowledge in accident appraisal. Thus, for the fair assessment of accident liabilities the correctness of these three key variables is of critical importance to police investigation reports.