Choosing a stock portfolio is always one of the most important issues for investors. Theoretically, selecting a stock portfolio can be solved by minimizing risk assumptions with the help of mathematical relationships, but with the variety of choices in the capital market, mathematical relationships alone are not an effective solution. The variety of investment tools and the differences in the functionality of investors’ complexity have complicated the selection process. Now the expansion of financial and capital markets, the use of rule-based systems for quick decisions, with minimal risk and away from human error, design, development, or improvement of these systems can be a competitive advantage. In the present study, neural network Algorithms and genetic programming Algorithms have been used to identify effective features and the decision tree to improve id3 has been proposed as a method for predicting price and trend of stock price change to select the optimal basket. The research results show that in addition to reducing computational and memory overhead, the proposed method is able to accurately predict severe fluctuations with nonlinear patterns and compared to modern methods such as nearest neighbor search, linear regression, autoregressive integrated moving average, and time series prophet algorithm will do better.

One of the most significant and effective criteria in the process of cutting dimensional rocks using the gang saw is the maximum energy consumption rate of the machine, and its accurate prediction and estimation can help designers and owners of this industry to achieve an optimal and economic process. In the present research work, it is attempted to study and provide models for predicting the maximum energy consumption of the gang saw during the process of soft dimensional rocks with the help of an intelligent optimization model such as random non-linear techniques, i. e. the Hybrid ANFIS-DE and Hybrid ANFIS-PSO Algorithms based upon 4 physical and mechanical parameters including uniaxial compressive strength, Mohs hardness, Schimazek’ s F-abrasiveness factors, Young modulus, and an operational characteristic of the machine, i. e. production rate. During this research work, 120 samples are tested on 12 carbonate rocks. The maximum energy consumption of the cutting machine during this work is measured and used as a modeling output for evaluating the performance of cutting machine. Also meta-heuristic Algorithms including DE and PSO Algorithms are used for training the Adaptive Neural Fuzzy Inference System (ANFIS). In addition, the PSO algorithm has a higher ability in terms of model output and performance indices and has a superiority over the differential evolution algorithm. Furthermore, comparison between the measured datasets with the ANFIS-DE and ANFIS-PSO models indicate the accuracy and ability of the ANFIS-PSO model in predicting the performance of gang saw considering the machine’ s properties and the cut rock.

In this paper Hybrid flowshop scheduling problem where some jobs, not all, have to follow no-wait restriction (that is, the operations of that job must be processed with no stop) is examined. In the literature, all papers assume that all jobs of the shops have to follow no-wait restrictions. First, this paper mathematically formulates the problem with two different mixed integer linear models under proposed considerations. The models are evaluated using two performance measures of size complexity and computational complexity. The small instances of the problem are solved using commercial software of mathematical programming. To solve larger instances of problem, two solution Algorithms have been developed. These two Algorithms are based on imperialist competitive algorithm and simulated annealing. A comprehensive numerical experiment including small and large instances is conducted to evaluate the models and Algorithms. The results show that the imperialist competitive algorithm outperforms simulated annealing.

The nonlinear Muskingum model has a significant advantage as compared to the linear model due to the nonlinear relationship between the storage and the flow dishcrage. In this model, the correct estimation of the parameters is necessary to achieve the proper precision. Previous studies indicated that there are five nonlinear corrected models which, with different optimization Algorithms, tried to increase the prediction accuracy of output hydrographs. Due to the error in the output hydrograph of the previous models, in this study, a new structure of nonlinear Muskingum model was developed based on Hybrid PSO and DSO Algorithms. In this eight-parameter model (NL6 model), the improvement coefficient γ was used which held values less or more than one according to the number of peak discharges in the output hydrograph. By applying the proposed approach to the three types of input hydrographs and determining the optimal values of the parameters for the NL6 model, this research showed that the proposed model has a high accuracy in estimating the discharge values of the output hydrograph. The error reduction rate of the NL6 model based on SSQ and SAD indicators for multi-peak hydrographs were 53 and 35.6 percent compared to the last proposed model, respectively. So, this model have a high performance in estimating flood routing hydrograph.

The widespread use of E-commerce websites has drastically increased the need for automatic recommendation systems with machine learning. In recent years, many ML-based recommenders and analysers have been built, however, their scope is limited to using a single filtering technique and processing with clustering-based predictions. This paper aims to provide a systematic year-wise survey and evolution of these existing recommenders and analysers in specific deep learning-based Hybrid filtering categories using movie datasets. They are compared to others based on their problem analysis, learning factors, data sets, performance, and limitations. Most contributions are found with collaborative filtering using user or item similarity and deep learning for the IMDB datasets. In this direction, this paper introduces a new and efficient Hybrid Filtering based Recommendation System using Deep Learning (HFRS-DL), which includes multiple layers and stages to provide a better solution for generating recommendations.

In this paper we have tried to develop an altered version of the artificial bee colony algorithm which is inspired from and combined with the meta-heuristic algorithm of firefly. In this method, we have tried to change the main equation of searching within the original ABC algorithm. On this basis, a new combined equation was used for steps of employed bees and onlooker bees. For this purpose, we had to define several new parameters for improving the quality of the proposed method. In this regard, we have introduced two new parameters to the method. The new method has been simulated within the software of MATLAB and it has also been run according to objective functions of SPHERE, GRIEWANK and ACKLEY. All these functions are standard evaluation functions that are generally used for meta-heuristic Algorithms. Results that were yielded by the proposed method were better than the results of the initial algorithm and especially by increasing the number of variables of the problem, this improvement becomes even more significant. We have successfully established a better balance between concepts of exploration and exploitation, especially with increasing the repetition cycles, we have successfully controlled the concept of utilization with random parameters. Tests have been ran more than 500 times.

Increase of stored data in medical databases needs allocative tools to get access to data, data mining, discover knowledge and efficient use of data. Medical and treatment fields are two examples of data mining tools to analyze massive data and predictive modelling. In medical sciences, prediction and precise-quick detection of multiple diseases has to reduced exprense and also save people’, s lives. Group based methods (Ensemble Methods) are approaches that use Hybrid models to recover classification. Coronavirus (COVID-19) has killed many people around the world so far, and this could be a good reason to present a new model for diagnosing the disease using data mining Algorithms. This research presents a Hybrid model of basic data mining and Hybrid Algorithms according to information in medical and laboratory records of patients suffering Covid-19 in Emam-Reza (AS) hospital in Mashhad, Iran, to diagnose the sickness. The proposed method uses Ensemble base (Hybrid) classifiers, where the general model can be used to provide diagnoses with higher precision rather than classifiers. To execute the proposed model, data mining tools including Rapid Miner 9. 7 and Python 3. 7 were used. This study used stacking classifiers composed of basic Algorithms including simple baze, decision tree, K-nearest neighborhood backup vector machine for basic section and uses chaos jungle algorithm in stack section that has gained 86. 5% accuracy for diagnosis of Covid-19.

In this paper we have tried to develop an altered version of the artificial bee colony algorithm which is inspired from and combined with the meta-heuristic algorithm of firefly. In this method, we have tried to change the main equation of searching within the original ABC algorithm. On this basis, a new combined equation was used for steps of employed bees and onlooker bees. For this purpose, we had to define several new parameters for improving the quality of the proposed method. In this regard, we have introduced two new parameters to the method. The new method has been simulated within the software of MATLAB and it has also been run according to objective functions of SPHERE, GRIEWANK and ACKLEY. All these functions are standard evaluation functions that are generally used for meta-heuristic Algorithms. Results that were yielded by the proposed method were better than the results of the initial algorithm and especially by increasing the number of variables of the problem, this improvement becomes even more significant.We have successfully established a better balance between concepts of exploration and exploitation, especially with increasing the repetition cycles, we have successfully controlled the concept of utilization with random parameters. Tests have been ran more than 500 times.

Based on the extragradient-like method combined with shrinking projection, we propose two Algorithms, the first algorithm is obtained using sequential computation of extragradient-like method and the second algorithm is obtained using parallel computation of extragradient-like method, to find a common point of the set of fixed points of a nonexpansive mapping and the solution set of the equilibrium problem of a bifunction given as a sum of the finite number of H ̈, older continuous bifunctions. The convergence theorems for iterative sequences generated by the Algorithms are established under widely used assumptions for the bifunction and its summands.