Whose Balance Sheet Is This? Neural Networks for Bankss Pattern Recognition

This study describes the implementation of Artificial Neural Network Pattern Recognition using Back-Propagation algorithm which can perform cardiovascular health pre-diagnosis of a patient through a generated blood pressure profile. The proponents gathered data from institutions that conduct Exercise Stress Testing, specifically the Treadmill Stress Test. The data gathered were age, gender, height, weight, blood pressure and heart rate readings and is considered as blood pressure condition risk factor. They compose the 47 input parameters of the network and was then divided into two — the training data and the testing data. This was put into a database created using Microsoft Excel. The back-propagation neural network model gives an accurate pre-diagnosis. The trained system gives an acceptable pre-diagnosis in reference to the given diagnosis by the attending practitioner that facilitated the collection of data.

A thin-film CMOS MEMS thermal sensor has been designed, fabricated and tested with the addition of through-membrane isolating holes. These holes have been shown to enhance the discrimination towards gases with differing thermal conductivity in the presence of flow. Using three on-membrane resistors as inputs, linear statistical methods alongside Artificial Neural Network pattern recognition techniques have been investigated for decoupling the two parameters of thermal conductivity and flow rate using a single sensor. In addition to this, the addition of the membrane holes increases the sensitivity towards flow rate by 10 times and the sensitivity towards thermal conductivity by 2 times. This sensor design coupled with well-known post-processing techniques will enable a new generation of multi-parameter sensing solutions.

The structural validation problem using quantum chemistry approaches (confirm or reject a candidate structure) has been tackled with artificial neural network (ANN) mediated multidimensional pattern recognition from experimental and calculated 2D C-H COSY. In order to identify subtle errors (such as regio- or stereochemical), more than 400 ANNs have been built and trained, and the most efficient in terms of classification ability were successfully validated in challenging real examples of natural product misassignments.

Error diagnosis and detection have become important in modern production due to the importance of spinning equipment. Artificial neural network pattern recognition methods are widely utilized in rotating equipment fault detection. These methods often need a large quantity of sample data to train the model; however, sample data (especially fault samples) are uncommon in engineering. Preliminary work focuses on dimensionality reduction for big data sets using semisupervised methods. The rotary machine’s polar coordinate signal is used to build a GAN network structure. ANN and tiny samples are utilized to identify DCGAN model flaws. The time-conditional generative adversarial network is proposed for one-dimensional vibration signal defect identification under data imbalance. Finally, auxiliary samples are gathered under similar conditions, and CCNs learn about target sample characteristics. Convolutional neural networks handle the problem of defect identification with small samples in different ways. In high-dimensional data sets with nonlinearities, low fault type recognition rates and fewer marked fault samples may be addressed using kernel semisupervised local Fisher discriminant analysis. The SELF method is used to build the optimum projection transformation matrix from the data set. The KNN classifier then learns low-dimensional features and detects an error kind. Because DCGAN training is unstable and the results are incorrect, an improved deep convolutional generative adversarial network (IDCGAN) is proposed. The tests indicate that the IDCGAN generates more real samples and solves the problem of defect identification in small samples. Time-conditional generation adversarial network data improvement lowers fault diagnosis effort and deep learning model complexity. The TCGAN and CNN are combined to provide superior fault detection under data imbalance. Modeling and experiments demonstrate TCGAN’s use and superiority.

Online detection technology research on agricultural product quality deterioration

Arcing fault identification using combined Gabor Transform-neural network for transmission lines

This article aimed to determine the effectiveness of (Machine Learning, Artificial Neural Networks, Pattern Recognition and Data Analysis, and Fire Dynamics Simulations) in improving the results of arson investigations in the United Arab Emirates (UAE). The research relied on the descriptive method. The study community consisted of all members of the teams specialized in fire investigations. The research sample was taken using a simple random method that included (74) individuals from the research community. The researcher relied on the questionnaire as the main instrument for collecting data from the study sample. The results of the study showed that artificial intelligence techniques (Machine Learning, Artificial Neural Networks, Pattern Recognition and Data Analysis, and Fire Dynamics Simulations) can be effective in improving the results of arson investigations. The study recommended providing training programs to investigators on how to make full use of AI techniques in the investigation process.

A pattern recognition artificial neural network method for random fatigue loading life prediction

Intelligent ground vehicles require automated steering control systems. The complexity of the automated steering systems has increased as the concurrent data receiving, processing, decision-making and monitoring all happen simultaneously in super-fast sampling rates. The research reported in this paper focuses on the neural network training and machine learning approach for an automated steering system. The essential information of the steering controller was trained using artificial neural network (ANN) and pattern recognition algorithm approaches. The objective is to investigate the design of ANN training and deep machine learning on collected data from sensors of steering wheel angular position, speed, steering column torque, and ego vehicle speed and generating acceptable steering commands for the vehicle. The results of the research showed that the proposed network control system had trained and validated more than 96.5% steering system behavior patterns and adapted large random disturbances of the steering controller commands. It is, therefore, necessary to develop artificial intelligence methodologies in automated steering systems of autonomous vehicles with neural network representing the main topology blocks of the control system architecture and utilize ANN abstraction in the control system of autonomous vehicle steering control system.

Artificial Neural Networks (ANN) are computational models inspired by the human brain, capable of recognizing patterns and making predictions. Scale Conjugate Gradient (SCG) algorithm is an efficient training method for ANN that accelerates the learning process and improves output accuracy. However, conventional ANN training methods often struggle with slow convergence and can be less accurate when analyzing complex, high-dimensional data such as Electroencephalogram (EEG) signals. Furthermore, the precise classification of subtle neural pattern changes induced by Radiofrequency (RF) exposure remains a significant challenge. SCG improves the learning process of ANNs by speeding up the adjustment of their internal weights, helping the network learn faster and more accurately from large data sets. This study aims to improve the classification of RF neural data patterns using SCG. EEG neural data was captured in sessions before, during and after RF exposure. Power Asymmetry Ratio (PAR) was used for feature extraction. The data involved 96 subjects, were split into 70:30 ratio for training and testing in ANN modelling. The SCG algorithm was integrated, initialized with one hidden layer of 10 neurons. Parameter adjustments were made to optimize convergence, potentially involving multiple layers for model refinement. The results show that RF exposure in During session produces significantly distinct neural patterns, enabling the highest ANN classification accuracy.

We investigate the usefulness of balance sheet disaggregations in the context of forecasting operating asset growth and its implications for revenue growth. We show that models using disaggregated balance sheet information have greater accuracy than models that use only aggregate information, with the most accurate disaggregation scheme separately considering property, plant, and equipment, intangible assets, other noncurrent operating assets, and current operating assets. When investigating market participants’ use of disaggregated balance sheet information, we find that differences in growth predictions from a disaggregated model and an aggregate model are associated with year-ahead abnormal returns, suggesting that investors do not fully utilize disaggregated balance sheet information. We corroborate this result by showing that returns are concentrated around the earnings announcement, are more pronounced for stocks with low transient institutional ownership, and are consistent with subsequent improvements in actual financial performance. Finally, we find that these returns are driven by both the asset growth predictions of a disaggregated model and the ability of such a model to incorporate the unique revenue-generating ability of individual asset components. Importantly, we find that our results are incremental to the information embedded in income statement disaggregations, even when forecasting income statement items such as profitability. Overall, our study provides evidence on the informativeness of balance sheet disaggregations and proposes the use of disaggregated models when forecasting growth in operating assets and its implications for future revenues. This paper was accepted by Eric So, accounting. Supplemental Material: The online appendix and data files are available at https://doi.org10.1287/mnsc.2023.01413 .

Integration of modern remote sensing technologies for faster utility mapping and data extraction

In case of a carbonate reservoir, conventional techniques such as artificial neural networks and pattern recognition may not perform well in predicting the reservoir potential zones. This paper proposes a new implementation to highlight the anomalous zones of a carbonate reservoir using grey theory. The input is a 3D seismic volume for a square area which contains eight drilled wells of varying producing quality. The grey analysis is carried out for the zone of interest on a trace-by-trace basis. The result is validated on the reservoir main producing layer by utilizing the wells production classification information.

Bankruptcy prediction is always a topical issue. The activities of all business entities are directly or indirectly affected by various external and internal factors that may influence a company in insolvency and lead to bankruptcy. It is important to find a suitable tool to assess the future development of any company in the market. The objective of this paper is to create a model for predicting potential bankruptcy of companies using suitable classification methods, namely Support Vector Machine and artificial neural networks, and to evaluate the results of the methods used. The data (balance sheets and profit and loss accounts) of industrial companies operating in the Czech Republic for the last 5 marketing years were used. For the application of classification methods, TIBCO’s Statistica software, version 13, is used. In total, 6 models were created and subsequently compared with each other, while the most successful one applicable in practice is the model determined by the neural structure 2.MLP 22-9-2. The model of Support Vector Machine shows a relatively high accuracy, but it is not applicable in the structure of correct classifications.

With the use of fuzzy logic techniques, neural computing can be integrated in symbolic reasoning to solve complex real world problems. In fact, artificial neural networks, expert systems, and fuzzy logic systems, in the context of approximate reasoning, share common features and techniques. A model of Fuzzy Connectionist Expert System is introduced, in which an artificial neural network is designed to construct the knowledge base of an expert system from, training examples (this model can also be used for specifications of rules in fuzzy logic control). Two types of weights are associated with the synaptic connections in an AND-OR structure: primary linguistic weights, interpreted as labels of fuzzy sets, and secondary numerical weights. Cell activation is computed through min-max fuzzy equations of the weights. Learning consists in finding the (numerical) weights and the network topology. This feedforward network is described and first illustrated in a biomedical application (medical diagnosis assistance from inflammatory-syndromes/proteins profiles). Then, it is shown how this methodology can be utilized for handwritten pattern recognition (characters play the role of diagnoses): in a fuzzy neuron describing a number for example, the linguistic weights represent fuzzy sets on cross-detecting lines and the numerical weights reflect the importance (or weakness) of connections between cross-detecting lines and characters.