This research develops a fuzzy-neural network model, using AI technologies and applies the model for effective control of profitability in paper recycling to improve production accuracy, reliability, robustness and to maximize profit generated by an industry, despite varying cost of production upon which ultimately profit, in an industry depend. Recycling reduces greenhouse gas emissions, conserves the natural resources on Earth, and saves space in the landfills for future generations of people. A sustainable future requires a high degree of recycling. However, Recycling industries face serious economic problems that increase the cost of recycling. Fuzzy logic has emerged as a tool to deal with uncertain, imprecise, partial truth or qualitative decisionmaking problems, to achieve robustness, tractability, and low cost, but it cannot automatically acquire the rules it uses to make those decisions. Neural networks have the ability to learn, generalize and process large amount of numerical data, but they are not good at explaining how they reach their decisions. The hybrid fuzzy-neural system has the ability to overcome the limitations of individual technique and enhances their strengths to handle financial trading. In order to achieve our objective, a study of a knowledge based system for effective control of profitability in paper recycling is carried out. The Mamdani's Max-Min technique is employed to infer data from the rules developed. This resulted in the establishment of some degrees of influence of input variables on the output. Fuzzy-Neural network model is developed using back propagation and supervised learning methods respectively. The outputs of Fuzzy logic serve as input to the neural network. To reinforce the proposed approach, we apply it to a case study performed on Paper recycling industry in Nigeria. A computer simulation is designed to assist the experimental decision for the best control action. The system is developed using MySQL, NetBeans, Java, MS Excel 2003, MatLab, etc. The obtained simulation and implementation fuzzy-neural results are investigated, compared and discussed.

The large nature of students’ dataset has made it difficult to find patterns associated with students’ academic performance (AP) using conventional methods. This has increased the rate of drop-outs, graduands with weak class of degree (CoD) and students that spend more than the minimum stipulated duration of studies. It is necessary to determine students’ AP using educational data mining (EDM) tools in order to know students who are likely to perform poorly at an early stage of their studies. This paper explores k-means and self-organizing map (SOM) in mining pieces of knowledge relating to the natural number of clusters in students’ dataset and the association of the input features using selected demographic, pre-admission and first year performance. Matlab 2015a was the programming environment and the dataset consists of nine sets of computer science graduands. Cluster validity assessment with k-means discovered four (4) clusters with correlation metric yielding the highest mean silhouette value of 0.5912.  SOM provided an hexagonal grid visual of feature component planes and scatter plots of each significant input attribute. The result shows that the significant attributes were highly correlated with each other except entry mode (EM), indicating that the impact of EM on CoD varies with students irrespective of mode of admission. Also, four distinct clusters were also discovered in the dataset by SOM —7.7% belonging to cluster 1 (first class), and 25% for cluster 2 (2nd class Upper) while Clusters 3 and 4 had 35% proportion each. This validates the results of k-means and further confirms the importance of early detection of students’ AP and confirms the effectiveness of SOM as a cluster validity tool. As further work, the labels from SOM will be associated with records in the dataset for association rule mining, supervised learning and prediction of students’ AP.

This paper proposes a fuzzy logic (FL) model for evaluation of quality of service (QoS) in multimedia transmission over ad hoc networks as an effective mechanism for QoS management. It aims at minimizing the negative effects of major QoS parameters, (jitter, delay, packet loss) sustaining efficiency and reliability of data deliveries and improving overall system performance. Both triangular membership function (TMF) and Gaussian membership function (GMF) are adopted to demonstrate their effects in FL-QoS evaluation. The proposed approaches are implemented in Matlab/Simulink. Results indicate that input conditions have varying level of influences on the output based on the MFs used. It is generally observed that QoS fuzzy control with TMF gives a better performance than GMF. Typically, it is also noted that when the input conditions are selected at 50[Formula: see text]ms delay, 5[Formula: see text]ms jitter and 50% packet loss, we obtain 79% and 71% output responses with TMF and GMF, respectively. This shows that TMF method can control QoS in a multimedia transmission over ad hoc wireless network more effectively with good service output response than GMF. Generally, the paper shows that FL is capable of measuring network performance and predicting any QoS deterioration without complex mathematical calculation, to provide an improved QoS for customers’ satisfaction.

This paper focuses on the study of short term load forecasting (STELF) using interval Type-2 Fuzzy Logic (IT2FL) and feed-forward Neural Network with back-propagation (NN-BP

Interval type-2 fuzzy logic systems (IT2FLSs), have recently shown great potential in various applications with dynamic uncertainties. It is believed that additional degree of uncertainty provided by IT2FL allows for better representation of the uncertainty and vagueness present in prediction models. However, determining the parameters of the membership functions of IT2FL is important for providing optimum performance of the system. Particle Swarm Optimization (PSO) has attracted the interest of researchers due to their simplicity, effectiveness and efficiency in solving real-world optimization problems. In this paper, a novel optimal IT2FLS is designed, applied for predicting winning chances in elections. PSO is used as an optimized algorithm to tune the parameter of the primary membership function of the IT2FL to improve the performance and increase the accuracy of the IT2F set. Simulation results show the superiority of the PSO-IT2FL to the similar non-optimal IT2FL system with an increase in the prediction.

No abstract

In this paper, a fuzzy expert system for the diagnosis and monitoring of cholera is presented for providing decision support platform to cholera researchers, physicians and other healthcare practitioners in cholera endemic regions. The developed fuzzy expert system composed of four components which include; the Knowledge base, the Fuzzification, the Inference engine and Defuzzification. Object oriented Design tools is adopted in the design of our database. We develop our knowledge based on clinical observations, medical diagnosis and the expert's knowledge. We employ Mamdani's MAX-MIN fuzzy inference engine to infer data from the rules developed. This resulted in the establishment of some degrees of influence of input variables on the output. The technique allows for mild, moderate and severe symptoms to be applied in order to get the estimation result. Triangular membership function is employed to evaluate the degree of participation of each input parameter and the defuzzification technique employed is the Centriod of Area (COA).Twenty patients with cholera are selected and studied and the observed results computed in the range of predefined limit by the domain experts. This system will offer potential assistance to medical practitioners and healthcare sector in making prompt decision during the diagnosis of cholera.

This study employed Support Vector Machine (SVM) in the classification and prediction of fire outbreak based on fire outbreak dataset captured from the Fire Outbreak Data Capture Device (FODCD). The fire outbreak data capture device (FODCD) used was developed to capture environmental parameters values used in this work. The FODCD device comprised DHT11 temperature sensor, MQ-2 smoke sensor, LM393 Flame sensor, and ESP8266 Wi-Fi module, connected to Arduino nano v3.0.board. 700 data point were captured using the FODCD device, with 60% of the dataset used for training while 20% was used for testing and validation respectively. The SVM model was evaluated using the True Positive Rate (TPR), False Positive Rate (FPR), Accuracy, Error Rate (ER), Precision, and Recall performance metrics. The performance results show that the SVM algorithm can predict cases of fire outbreak with an accuracy of 80% and a minimal error rate of 0.2%. This system was able to predict cases of fire outbreak with a higher degree of accuracy. It is indicated that the use of sensors to capture real world dataset, and machine learning algorithm such as support vector machine gives a better result to the problem of fire management.