International Journal of Intelligent Systems and Applications @ijisa
Статьи журнала - International Journal of Intelligent Systems and Applications
Все статьи: 1187
Predicting Student Academic Performance at Degree Level: A Case Study
Статья научная
Universities gather large volumes of data with reference to their students in electronic form. The advances in the data mining field make it possible to mine these educational data and find information that allow for innovative ways of supporting both teachers and students. This paper presents a case study on predicting performance of students at the end of a university degree at an early stage of the degree program, in order to help universities not only to focus more on bright students but also to initially identify students with low academic achievement and find ways to support them. The data of four academic cohorts comprising 347 undergraduate students have been mined with different classifiers. The results show that it is possible to predict the graduation performance in 4th year at university using only pre-university marks and marks of 1st and 2nd year courses, no socio-economic or demographic features, with a reasonable accuracy. Furthermore courses that are indicators of particularly good or poor performance have been identified.
Бесплатно
Prediction of Adsorption of Cadmium by Hematite Using Fuzzy C-Means Clustering Technique
Статья научная
Clustering is partitioning of data set into subsets (clusters), so that the data in each subset share some common trait. In this paper, an algorithm has been proposed based on Fuzzy C-means clustering technique for prediction of adsorption of cadmium by hematite. The original data elements have been used for clustering the random data set. The random data have been generated within the minimum and maximum value of test data. The proposed algorithm has been applied on random dataset considering the original data set as initial cluster center. A threshold value has been taken to make the boundary around the clustering center. Finally, after execution of algorithm, modified cluster centers have been computed based on each initial cluster center. The modified cluster centers have been treated as predicted data set. The algorithm has been tested in prediction of adsorption of cadmium by hematite. The error has been calculated between the original data and predicted data. It has been observed that the proposed algorithm has given better result than the previous applied methods.
Бесплатно
Prediction of Drought Resistance Gene with Clustered Amino Acid Features
Статья научная
Drought resistant gene plays important role in molecular breeding while little is known for its genetic mechanism. By extracting the clustered amino acids features, crucial numerical features are inferred for the resistance property of the given gene. Support vector machine algorithm is used to testify the reliability of feature extraction method. After carefully parameters choosing, the accuracy of the predictor achieves 79.36% in Jack-knife test, and the Mathews correlation coefficient achieves 0.5636.
Бесплатно
Prediction of Missing Associations Using Rough Computing and Bayesian Classification
Статья научная
Information technology revolution has brought a radical change in the way data are collected or generated for ease of decision making. It is generally observed that the data has not been consistently collected. The huge amount of data has no relevance unless it provides certain useful information. Only by unlocking the hidden data we can not use it to gain insight into customers, markets, and even to setup a new business. Therefore, the absence of associations in the attribute values may have information to predict the decision for our own business or to setup a new business. Based on decision theory, in the past many mathematical models such as naïve Bayes structure, human composed network structure, Bayesian network modeling etc. were developed. But, many such models have failed to include important aspects of classification. Therefore, an effort has been made to process inconsistencies in data being considered by Pawlak with the introduction of rough set theory. In this paper, we use two processes such as pre process and post process to predict the output values for the missing associations in the attribute values. In pre process we use rough computing, whereas in post process we use Bayesian classification to explore the output value for the missing associations and to get better knowledge affecting the decision making.
Бесплатно
Prediction of Operating Loads Contribution to Assembly Relation and Product Behavior
Статья научная
In the process of product manufacturing, control of assembly error will directly affect product operating behavior. When product running, operating loads will lead to change of assembly relation of product parts, affecting product behavior. Based on Jacobian-Torsor method, the Jacobian Torsor tolerance model, considering contribution of operating loads, was extended and corrected, the assembly error (assembly relation change) resulted from operating loads can be calculated. Variation of running behavior with assembly error was divided to three phases: compensation phase, rapid loss phase and total loss phase. Based on changing curve of product behavior, function of behavior loss was constructed to describe behavior loss resulting from assembly error of a part of product. The conception and calculating method of behavior loss index (BLI) are given to reflect behavior changing status of whole product under certain assembly accuracy. Combined with extended Jacobia -Torsor method, the calculated results can be used to predict product behavior change led by operating loads. The prediction can help to know next measurement adopted in product design phase. An example is given to demonstrate calculating procedure of given method.
Бесплатно
Prediction of Possible Business of a Newly Launched Film using Ordinal Values of Film-genres
Статья научная
Film industry is the most important component of entertainment industry. Both profit and loss are very high for this business. Like every other business, business prediction system plays a vital role for this industry. Before release of a particular movie, if the Production Houses or distributors get any type of prediction that how the film will do business, then it will be very useful to reduce the risk of the investors. In this paper we have proposed a method using back propagation neural network for prediction about a given movie’s profitability. Initially the entire range of profit-loss has been divided into a number of groups. The proposed algorithm can assign a given movie to it’s appropriate profit-loss group. Note that, a similar such method has been successfully applied in the field of Stock Market Prediction, Weather Prediction and Image Processing.
Бесплатно
Prediction of Rainfall in India using Artificial Neural Network (ANN) Models
Статья научная
In this paper, ARIMA(1,1,1) model and Artificial Neural Network (ANN) models like Multi Layer Perceptron (MLP), Functional-link Artificial Neural Network (FLANN) and Legendre Polynomial Equation ( LPE) were used to predict the time series data. MLP, FLANN and LPE gave very accurate results for complex time series model. All the Artificial Neural Network model results matched closely with the ARIMA(1,1,1) model with minimum Absolute Average Percentage Error(AAPE). Comparing the different ANN models for time series analysis, it was found that FLANN gives better prediction results as compared to ARIMA model with less Absolute Average Percentage Error (AAPE) for the measured rainfall data.
Бесплатно
Prediction of Stock Market in Nigeria Using Artificial Neural Network
Статья научная
Prediction of Nigerian stock market is almost not done by any researcher and is an important factor which can be used to determine the viability of Nigerian stock market. In this paper, the prediction models were developed using Artificial Neural Network. The result of the prediction of Nigerian Stock Exchange (NSE) market index value of selected banks using Artificial Neural Network was presented. The multi-layer feed forward neural network was used, so that each output unit is told what its desired response to input signals ought to be. This work has confirmed the fact that artificial neural network can be used to predict future stock prices. The data collection period is from 2003 to 2006.
Бесплатно
Prediction of performance point of semi-rigid steel frames using artificial neural networks
Статья научная
One of the main steps in the performance based seismic analysis and design of structures is determination of performance point where the nonlinear static analysis approach is used. The aim of this paper is to predict the performance point of semi-rigid steel frames using Artificial Neural Networks. As such, to generate data required for the prediction, several semi-rigid steel frames were modeled and their performance point was determined then. Ten input variables including number of bays, number of stories, bays width, moment of inertia of beams, cross sectional area of columns, cross sectional area of braces, rigidity degree of connections and soft story (existence or nonexistence) were considered in the prediction. In addition, the actual results were obtained at the presence of different earthquake intensity levels and soil types. Back Propagation with eleven different algorithms and Radial Basis Function Artificial Neural Networks were used in the prediction. The prediction process was carried out in two steps. In the first step, all samples were used for the prediction and the performance metrics were computed. In the second step, three of the best networks were selected, and the optimum number of samples was found considering a very slight reduction in the accuracy of the networks used. Finally, it was shown that, despite using rather limited number of samples, the generated Artificial Neural Networks accurately predict the performance point of semi-rigid steel frames.
Бесплатно
Prediction of water demand using artificial neural networks models and statistical model
Статья научная
The prediction of future water demand will help water distribution companies and government to plan the distribution process of water, which impacts on sustainable development planning. In this paper, we use a linear and nonlinear models to predict water demand, for this purpose, we will use different types of Artificial Neural Networks (ANNs) with different learning approaches to predict the water demand, compared with a known type of statistical methods. The dataset depends on sets of collected data (extracted from municipalities databases) during a specific period of time and hence we proposing a nonlinear model for predicting the monthly water demand and finally provide the more accurate prediction model compared with other linear and nonlinear methods. The applied models capable of making an accurate prediction for water demand in the future for the Jenin city at the north of Palestine. This prediction is made with a time horizon month, depending on the extracted data, this data will be used to feed the neural network model to implement mechanisms and system that can be employed to predicts a short-term for water demands. Two applied models of artificial neural networks are used; Multilayer Perceptron NNs (MLPNNs) and Radial Basis Function NNs (RBFNNs) with different learning and optimization algorithms Levenberg Marquardt (LM) and Genetic Algorithms (GAs), and one type of linear statistical method called Autoregressive integrated moving average ARIMA are applied to the water demand data collected from Jenin city to predict the water demand in the future. The execution results appear that the MLPNNs-LM type is outperformed the RBFNN-GAs and ARIMA models in the prediction the water demand values.
Бесплатно
Process Modeling and Simulation of Feedwater Heaters Drains and Vents System of PFBR
Статья научная
Nuclear Power Plants are a complex system and need to be controlled very meticulously to avoid any catastrophe from occurring. The safety and availability of the power plant relies on the human operators both through their ability and reliability to ensure smooth and trouble-free plant operations. Training the operators on normal plant operation, maintenance, fault diagnosis and unforeseen emergencies in the plant helps reduce the latency period of the plant and thus increase the efficiency. Operator Training Simulator has become an indispensable entity in imparting hands on training to these operators. Development of process simulators calls for the process to be designed, modeled and implemented to replicate the real plant in steady state and transient conditions.
Бесплатно
Producer-Scrounger Method to Solve Traveling Salesman Problem
Статья научная
Algorithms inspired from natural phenomena are seem to be efficient to solve various optimization problems. This paper investigates a new technique inspiring from the animal group living behavior to solve traveling salesman problem (TSP), the most popular combinatorial optimization problem. The proposed producer-scrounger method (PSM) models roles and interactions of three types of animal group members: producer, scrounger and dispersed. PSM considers a producer having the best tour, few dispersed members having worse tours and scroungers. In each iteration, the producer scans for better tour, scroungers explore new tours while moving toward producer’s tour; and dispersed members randomly checks new tours. For producer’s scanning, PSM randomly selects a city from the producer’s tour and rearranges its connection with several near cities for better tours. Swap operator and swap sequence based operation is employed in PSM to update a scrounger towards the producer. The proposed PSM has been tested on a large number of benchmark TSPs and outcomes compared to genetic algorithm and ant colony optimization. Experimental results revealed that proposed PSM is a good technique to solve TSP providing the best tours in most of the TSPs.
Бесплатно
Product Defect Detection Using Deep Learning
Статья научная
To maximize production efficiency, product quality control is paying more attention to the quick and reliable automated quality visual inspection. Product defect detection is a critical part of the inspection process. Manual defect detection has a lot of flaws that can be overcome using a deep learning approach. In this paper we have proposed and implemented the deep learning models to detect defects in the manufactured product. Two types of classification, i.e., binary and multiclass classification, is done using CNN, AlexNet, and YOLO algorithms. For the binary classification which is just used to check whether there is a defect in the product, we have proposed three different architectures of CNN, out of which the third CNN model gave 99.44% and 97.49% for training and testing, respectively. We also tested the AlexNet model and got accuracy of 97.6%. And for the multiclass classification that is used for identification of type(s) of defects, the YOLOv8 model is proposed and implemented, which gives better results by attaining a remarkable accuracy of 98.7% for multiclass classification. We also designed and developed the Android Application, which is used on the field for defect detection in the manufacturing industry.
Бесплатно
Proficient D-SEP Protocol with Heterogeneity for Maximizing the Lifetime of Wireless Sensor Networks
Статья научная
In this paper, we have proposed a new SEP protocol called as Deterministic-SEP (D-SEP), for electing cluster heads in a distributed fashion in two-, three-, and multi-level hierarchical wireless sensor networks. The significant improvement has been shown using D-SEP in comparison with SEP in terms of network lifetime, energy consumption and data transmission to BS. Our expectations are demonstrated by simulation results. We have introduced the superior characteristic of our protocol and discussed the cluster head selection algorithm by describing the threshold and probability equations. In order to reach the constructive conclusion, two cases of two-level and four cases of three-level heterogeneity have been reported and compared. The results reveal that there is 323% & 207% improvement in the overall lifetime of the network by using D-SEP after comparing two-level (m=0.3, a=1.5) & three-level (m=0.5, m0=0.4, a=1.5, b=3) respectively. The investigations ascertain the stable region and maximized lifetime of the network by using D-SEP over SEP. The development of 17.8 fold in the lifetime of the network is reported by using D-SEP. Moreover the energy depletion slope per round is lower in case of D-SEP over SEP.
Бесплатно
Статья научная
Basic rough set model introduced by Pawlak in 1982 has been extended in many directions to enhance their modeling power. One such attempt is the notion of rough sets on fuzzy approximation spaces by De et al in 1999. This basic model uses equivalence relation for its definition, which decompose the universal set into disjoint equivalence classes. These equivalence classes are called granules of knowledge. From the granular computing point of view the basic rough set model is unigranular in character. So, in order to handle more than one granular structure simultaneously, two types of multigranular rough sets, called the optimistic and pessimistic multigranular rough sets were introduced by Qian et al in 2006 and 2010 respectively. In this paper, we introduce two types of multigranular rough sets on fuzzy approximation spaces (optimistic and pessimistic), study several of their properties and illustrate how this notion can be used for prediction of rainfall. The introduced notions are explained through several examples.
Бесплатно
Proposed Representation Approach Based on Description Logics Formalism
Статья научная
The most familiar concept in Artificial intelligence is the knowledges representation. It aims to find explicit symbolization covering all semantic aspects of knowledge, and to make possible the use of this representation to produce an intelligent behavior like reasoning. The most important constraint is the usability of the representation; it's why the structures used must be well defined to facilitate manipulation for reasoning algorithms which leads to facilitate their implementation. In this paper we propose a new approach based on the description logics formalism for the goal of simplification of description logics system implementation. This approach can reduce the complexity of reasoning Algorithm by the vectorisation of concept definition based on the subsumption hierarchy.
Бесплатно
Proposing Two Defuzzification Methods based on Output Fuzzy Set Weights
Статья научная
Defuzzification converts the final fuzzy output set of fuzzy controller and fuzzy inference systems to a significant crisp value. However, there are various mathematical methods for defuzzification, but there is not any certain systematic method for choosing the best strategy. In this paper, first we explain the structure of a fuzzy inference system and then after a short review of defuzzification criteria and properties, the main classification groups of most widely used defuzzification methods are presented. In the following after discussing some existing techniques, two new defuzzification methods are proposed by presenting their general performance and computational formulas. However, the principle of these two methods is using weights associated with output fuzzy set like WFM or QM, but unlike the existing approaches, they consider the final aggregated consequent and implicated functions simultaneously to calculate the weights. To show how the proposed methods act, two numerical examples are solved using the presented methods and the results are compared with some of common defuzzification techniques.
Бесплатно
Статья научная
Recently, series compensation is widely used in transmission. However, this creates several problems to conventional protection approaches. This paper presents overcurrent and distance protection schemes, for fault classification in transmission lines with thyristor controlled series capacitor (TCSC) using support vector machine (SVM). The fault classification task is divided into four separate subtasks (SVMa, SVMb, SVMc and SVMg), where the state of each phase and ground is determined by an individual SVM. The polynomial kernel SVM is designed to provide the optimal classification conditions. Wide variations of load angle, fault inception angle, fault resistance and fault location have been carried out with different types of faults using PSCAD/EMTDC program. Backward faults have also been included in the data sets. The proposed technique is tested and the results verify its fastness, accuracy and robustness.
Бесплатно
QoS Metrics for Cloud Computing Services Evaluation
Статья научная
Cloud systems are transforming the Information Technology trade by facultative the companies to provide admission to their structure and also software products to the membership foundation. Because of the vast range within the delivered Cloud solutions, from the customer’s perspective of an aspect, it's emerged as troublesome to decide whose providers they need to utilize and then what's the thought of his or her option. Especially, employing suitable metrics is vital in assessing practices. Nevertheless, to the most popular of our knowledge, there's no methodical explanation relating to metrics for estimating Cloud products and services. QoS (Quality of Service) metrics playing an important role in selecting Cloud providers and also optimizing resource utilization efficiency. While many reports have got to devote to exploitation QoS metrics, relatively not much equipment supports the remark and investigation of QoS metrics of Cloud programs. To guarantee a specialized product is published, describing metrics for assessing the QoS might be an essential necessity. So, this text suggests various QoS metrics for service vendors, especially thinking about the consumer’s worry. This article provides the metrics list may stand to help the future study and also assessment within the field of Cloud service's evaluation.
Бесплатно
Qualitative and Quantitative Evaluation of EEG Signals in Epileptic Seizure Recognition
Статья научная
A chaos-ANFIS approach is presented for analysis of EEG signals for epileptic seizure recognition. The non-linear dynamics of the original EEGs are quantified in the form of the hurst exponent (H) and largest lyapunov exponent (λ). The process of EEG analysis consists of two phases, namely the qualitative and quantitative analysis. The classification ability of the H and λ measures is tested using ANFIS classifier. This method is evaluated with using a benchmark EEG dataset, and qualitative and quantitative results are presented. Our inter-ictal EEG based diagnostic approach achieves 97.4% accuracy with using 4-fold cross validation. Diagnosis based on ictal data is also tested in ANFIS classifier, reaching 96.9% accuracy. Therefore, our method can be successfully applied to both inter-ictal and ictal data.
Бесплатно
