Search results

This paper presents a combined method of ensemble learning and genetics to rebalance the corporate portfolio. The primary purpose of this paper is to determine the amount of investment in each of the shares of the listed company and the time of purchase, holding or sale of shares to maximize total return and reduce investment risk.

To achieve the goals of the problem, a two-level combined intelligent method, such as a support vector machine, decision tree, network Bayesian, k-nearest neighbors and multilayer perceptron neural network as heterogeneous basic models of ensemble learning in the first level, was applied. Then, the majority vote method (weighted average) in the second stage as the final model of learning was collectively used. Therefore, the data collected from 208 listed companies active in the Tehran stock exchange (http://tsetmc.com) from 2011 to 2015 have been used to teach the data. For testing and analysis, the data of the same companies between 2016 and 2020 have been used.

The results showed that the method of combined ensemble learning and genetics has the highest total stock portfolio yield of 114.12%, with a risk of 0.905%. Also, by examining the rate of return on capital, it was observed that the proposed method has the highest average rate of return on investment of 110.64%. As a result, the proposed method leads to higher returns with lower risk than the purchase and maintenance method for fund managers and companies and predicts market trends.

In the forthcoming research, there were no limitations to obtain research data were easily extracted from the site of Tehran Stock Exchange Technology Management Company and Rahvard Novin software, and simulation was performed in MATLAB software.

In this paper, using combined machine learning methods, companies’ stock prices are predicted and stock portfolio optimization is optimized. As companies and private organizations are trying to increase their rate of return, so they need a way to predict stock prices based on specific indicators. It turned out that this algorithm has the highest stock portfolio return with reasonable investment risk, and therefore, investors, portfolio managers and market timers can be used this method to optimize the stock portfolio.

The homogeneous and heterogeneous two-level hybrid model presented in the research can be used to predict market trends by market timers and fund managers. Also, adjusting the portfolio with this method has a much higher return than the return on buying and holding, and with controlled risk, it increases the security of investors’ capital, and investors invest their capital in the funds more safely. And will achieve their expected returns. As a result, the psychological security gained from using this method for portfolio arrangement will eventually lead to the growth of the capital market.

This paper tries to present the best combination of stock portfolios of active companies of the Tehran Stock Exchange by using the two-level combined intelligent method and genetic algorithm.

Solar power forecasting will have a significant impact on the future of large-scale renewable energy plants. Predicting photovoltaic power generation depends heavily on climate conditions, which fluctuate over time. In this research, we propose a hybrid model that combines machine-learning methods with Theta statistical method for more accurate prediction of future solar power generation from renewable energy plants. The machine learning models include long short-term memory (LSTM), gate recurrent unit (GRU), AutoEncoder LSTM (Auto-LSTM) and a newly proposed Auto-GRU. To enhance the accuracy of the proposed Machine learning and Statistical Hybrid Model (MLSHM), we employ two diversity techniques, i.e. structural diversity and data diversity. To combine the prediction of the ensemble members in the proposed MLSHM, we exploit four combining methods: simple averaging approach, weighted averaging using linear approach and using non-linear approach, and combination through variance using inverse approach. The proposed MLSHM scheme was validated on two real-time series datasets, that sre Shagaya in Kuwait and Cocoa in the USA. The experiments show that the proposed MLSHM, using all the combination methods, achieved higher accuracy compared to the prediction of the traditional individual models. Results demonstrate that a hybrid model combining machine-learning methods with statistical method outperformed a hybrid model that only combines machine-learning models without statistical method.

This study conducts a comparative study on the performance of reliability assessment methods based on adaptive surrogate models to accurately assess the reliability of automobile components, which is critical to the safe operation of vehicles.

In this study, different adaptive learning strategies and surrogate models are combined to study their performance in reliability assessment of automobile components.

By comparing the reliability evaluation problems of four automobile components, the Kriging model and Polynomial Chaos-Kriging (PCK) have better robustness. Considering the trade-off between accuracy and efficiency, PCK is optimal. The Constrained Min-Max (CMM) learning function only depends on sample information, so it is suitable for most surrogate models. In the four calculation examples, the performance of the combination of CMM and PCK is relatively good. Thus, it is recommended for reliability evaluation problems of automobile components.

Although a lot of research has been conducted on adaptive surrogate-model-based reliability evaluation method, there are still relatively few studies on the comprehensive application of this method to the reliability evaluation of automobile component. In this study, different adaptive learning strategies and surrogate models are combined to study their performance in reliability assessment of automobile components. Specially, a superior surrogate-model-based reliability evaluation method combination is illustrated in this study, which is instructive for adaptive surrogate-model-based reliability analysis in the reliability evaluation problem of automobile components.

The purpose of this paper is to report a new kind of workshop process which aims at co‐creation across disciplines in a service research network. The case concerns Technical Research Centre of Finland (VTT) and took place from January to May, 2009.

Both foresight and organizational learning methods are combined in the process. During workshops, researchers and management are enabled to co‐create interdisciplinary service research proposals and a service research strategy for VTT. The workshops are designed to facilitate a dialogue between users of the research and potential collaborators (universities, funding agencies and societal actors). This initiative reflects the current global service science discourse based on a renewal of service management through service‐dominant logic and network thinking.

Although the need for co‐creation across disciplines and together with the customer has often been stated in service research, methods enabling such a way of acting have rarely been tested and achieved. This method worked as a concrete way for managing future‐oriented networking across organizational borders as a basis for continuous learning and innovation.

The new approach to service science and the methods used in the VTT network are applicable in research practice.

The development process presented in this paper is an embryo for a new kind of research culture that fosters learning in networks as well as the shared and transparent planning of project proposals.

By creating the service science and business network and a process of learning by foresighting and evaluating our ideas on a concrete case are applied. This is believed to be the first time that methods of foresight and organizational learning have been combined. Furthermore, the process builds a research strategy both from below and above and together with customers and other collaborators thus establishing a network of co‐creation.

This paper aims to realize a fully distributed multi-UAV collision detection and avoidance based on deep reinforcement learning (DRL). To deal with the problem of low sample efficiency in DRL and speed up the training. To improve the applicability and reliability of the DRL-based approach in multi-UAV control problems.

In this paper, a fully distributed collision detection and avoidance approach for multi-UAV based on DRL is proposed. A method that integrates human experience into policy training via a human experience-based adviser is proposed. The authors propose a hybrid control method which combines the learning-based policy with traditional model-based control. Extensive experiments including simulations, real flights and comparative experiments are conducted to evaluate the performance of the approach.

A fully distributed multi-UAV collision detection and avoidance method based on DRL is realized. The reward curve shows that the training process when integrating human experience is significantly accelerated and the mean episode reward is higher than the pure DRL method. The experimental results show that the DRL method with human experience integration has a significant improvement than the pure DRL method for multi-UAV collision detection and avoidance. Moreover, the safer flight brought by the hybrid control method has also been validated.

The fully distributed architecture is suitable for large-scale unmanned aerial vehicle (UAV) swarms and real applications. The DRL method with human experience integration has significantly accelerated the training compared to the pure DRL method. The proposed hybrid control strategy makes up for the shortcomings of two-dimensional light detection and ranging and other puzzles in applications.

This paper presents an educational virtual reality (VR) game and experiments with different methods of including it into the teaching process. The purpose of this research study is to discover if immersive VR games can be used as an effective pedagogical tool if blended with traditional lectures by assisting learning gain, memory and knowledge retention while increasing edutainment value.

This research design comprises three different methods of learning: lecture-based involving lecture slides, infographics, and a video, game-based involving an immersive VR game of oil rig exploration, and the combination of lecture and game-based. Participants of each method filled up a questionnaire before and after participation to measure the learning gain, memory, and knowledge retention.

From the predominant findings of the study, the combined method demonstrated a significant increase in learning gain, memory, and knowledge retention and maybe a potentially suitable pedagogical tool.

Limitations of the study include findings based on one VR game with a specific educational topic, additionally, it is suspected that having different participants for each of the three methods may have slightly affected the results, albeit to a limited extent.

Findings of this study will provide evidence that VR games can be used alongside traditional lectures to aid in the learning process. Educators can choose to include VR games into their curriculums to improve the educational delivery process.

This research contributes to ways of incorporating VR games into educational curriculums through findings of this study highlighting the combination of VR games with lectures.

This study seeks to understand the connection of methodology by finding relevant papers and their full review using the “Preferred Reporting Items for Systematic Reviews and Meta-Analyses” (PRISMA).

Concrete-filled steel tubular (CFST) columns have gained popularity in construction in recent decades as they offer the benefit of constituent materials and cost-effectiveness. Artificial Neural Networks (ANNs), Support Vector Machines (SVMs), Gene Expression Programming (GEP) and Decision Trees (DTs) are some of the approaches that have been widely used in recent decades in structural engineering to construct predictive models, resulting in effective and accurate decision making. Despite the fact that there are numerous research studies on the various parameters that influence the axial compression capacity (ACC) of CFST columns, there is no systematic review of these Machine Learning methods.

The implications of a variety of structural characteristics on machine learning performance parameters are addressed and reviewed. The comparison analysis of current design codes and machine learning tools to predict the performance of CFST columns is summarized. The discussion results indicate that machine learning tools better understand complex datasets and intricate testing designs.

This study examines machine learning techniques for forecasting the axial bearing capacity of concrete-filled steel tubular (CFST) columns. This paper also highlights the drawbacks of utilizing existing techniques to build CFST columns, and the benefits of Machine Learning approaches over them. This article attempts to introduce beginners and experienced professionals to various research trajectories.

This paper presents a set of instrumental case studies for the context-based learning of BIM in the milieu of knowledge-based practice in the AEC industry. The study aimed to examine students' actions and perspectives in a simulated learning environment for real-world BIM processes. The core intent was to provide an in-depth understanding of strategic and functional BIM implementation by synthesizing a suggestive pedagogical framework based on context-based learning approaches.

Derived from context-based approaches and experiential learning methods such as role-play, problem-based and active learning, the study involved a set of doctoral-level case studies. In a qualitative research study, these cases were devised and organized around industry-focused simulations on various levels of BIM implementation strategies.

Results from the case studies and the student responses suggest that the comprehensive evaluation of real-world BIM implementation simulations facilitates a solid understanding of the value of BIM. The participation of industry professionals catalyzes the development of strategic and functional BIM competencies.

The study proposes a well-structured and replicable BIM learning framework based on context-based learning approaches. The novel framework is adaptive and flexible for BIM education. It can provide students with the necessary skills, strategic vision and professional competencies for innovative practices in the 21st-century AEC Industry. The simulative learning settings, including the evaluation rubrics and connected instructional methods, can be implemented and further developed for similar education efforts.

The individual machine learning methods used for fault detection and classification have accuracy performance at a certain level. A combined learning model composed of different base classifiers rather than an individual machine learning model is introduced to ensure diversity. In this way, this study aims to improve the generalization capability of fault detection and classification scheme.

This study presents a probabilistic weighted voting model (PWVM) with multiple learning models for fault detection and classification. The working principle of this study’s proposed model relies on weight selection and per-class possibilities corresponding to predictions of base classifiers. Moreover, it can improve the power of the prediction model and cope with imbalanced class distribution through validation metrics and F-score.

The performance of the proposed PWVM was better than the performance of the individual machine learning methods. Besides, the proposed voting model’s performance was compared with different voting mechanisms involving weighted and unweighted voting models. It can be seen from the results that the presented model is superior to voting mechanisms. The performance results revealed PWVM has a powerful predictive model even in noisy conditions. This study determines the optimal model from among voting models with the prioritization method on data sets partitioned different ratios. The obtained results with statistical analysis verified the validity of the proposed model. Besides, the comparative results from different benchmark data sets verified the effectiveness and robustness of this study’s proposed model.

The contribution of this study is that PWVM is an ensemble model with outstanding generalization capability. To the best of the authors’ knowledge, no study has been performed using a PWVM composed of multiple classifiers to detect no-faulted/faulted cases and classify faulted phases.

Hundreds of thousands of deaths each year in the world are caused by breast cancer (BC). An early-stage diagnosis of this disease can positively reduce the morbidity and mortality rate by helping to select the most appropriate treatment options, especially by using histological BC images for the diagnosis.

The present study proposes and evaluates a novel approach which consists of 24 deep hybrid heterogenous ensembles that combine the strength of seven deep learning techniques (DenseNet 201, Inception V3, VGG16, VGG19, Inception-ResNet-V3, MobileNet V2 and ResNet 50) for feature extraction and four well-known classifiers (multi-layer perceptron, support vector machines, K-nearest neighbors and decision tree) by means of hard and weighted voting combination methods for histological classification of BC medical image. Furthermore, the best deep hybrid heterogenous ensembles were compared to the deep stacked ensembles to determine the best strategy to design the deep ensemble methods. The empirical evaluations used four classification performance criteria (accuracy, sensitivity, precision and F1-score), fivefold cross-validation, Scott–Knott (SK) statistical test and Borda count voting method. All empirical evaluations were assessed using four performance measures, including accuracy, precision, recall and F1-score, and were over the histological BreakHis public dataset with four magnification factors (40×, 100×, 200× and 400×). SK statistical test and Borda count were also used to cluster the designed techniques and rank the techniques belonging to the best SK cluster, respectively.

Results showed that the deep hybrid heterogenous ensembles outperformed both their singles and the deep stacked ensembles and reached the accuracy values of 96.3, 95.6, 96.3 and 94 per cent across the four magnification factors 40×, 100×, 200× and 400×, respectively.

The proposed deep hybrid heterogenous ensembles can be applied for the BC diagnosis to assist pathologists in reducing the missed diagnoses and proposing adequate treatments for the patients.