Search results

With the development of intelligent technology, deep learning has made significant progress and has been widely used in various fields. Deep learning is data-driven, and its training process requires a large amount of data to improve model performance. However, labeled data is expensive and not readily available.

To address the above problem, researchers have integrated semi-supervised and deep learning, using a limited number of labeled data and many unlabeled data to train models. In this paper, Generative Adversarial Networks (GANs) are analyzed as an entry point. Firstly, we discuss the current research on GANs in image super-resolution applications, including supervised, unsupervised, and semi-supervised learning approaches. Secondly, based on semi-supervised learning, different optimization methods are introduced as an example of image classification. Eventually, experimental comparisons and analyses of existing semi-supervised optimization methods based on GANs will be performed.

Following the analysis of the selected studies, we summarize the problems that existed during the research process and propose future research directions.

This paper reviews and analyzes research on generative adversarial networks for image super-resolution and classification from various learning approaches. The comparative analysis of experimental results on current semi-supervised GAN optimizations is performed to provide a reference for further research.

By introducing Stacking algorithm to solve the underfitting problem caused by insufficient data in traditional machine learning, this paper provides a new solution to the cold start problem of entrepreneurial borrowing risk control.

The authors introduce semi-supervised learning and integrated learning into the field of migration learning, and innovatively propose the Stacking model migration learning, which can independently train models on entrepreneurial borrowing credit data, and then use the migration strategy itself as the learning object, and use the Stacking algorithm to combine the prediction results of the source domain model and the target domain model.

The effectiveness of the two migration learning models is evaluated with real data from an entrepreneurial borrowing. The algorithmic performance of the Stacking-based model migration learning is further improved compared to the benchmark model without migration learning techniques, with the model area under curve value rising to 0.8. Comparing the two migration learning models reveals that the model-based migration learning approach performs better. The reason for this is that the sample-based migration learning approach only eliminates the noisy samples that are relatively less similar to the entrepreneurial borrowing data. However, the calculation of similarity and the weighing of similarity are subjective, and there is no unified judgment standard and operation method, so there is no guarantee that the retained traditional credit samples have the same sample distribution and feature structure as the entrepreneurial borrowing data.

From a practical standpoint, on the one hand, it provides a new solution to the cold start problem of entrepreneurial borrowing risk control. The small number of labeled high-quality samples cannot support the learning and deployment of big data risk control models, which is the cold start problem of the entrepreneurial borrowing risk control system. By extending the training sample set with auxiliary domain data through suitable migration learning methods, the prediction performance of the model can be improved to a certain extent and more generalized laws can be learned.

This paper introduces the thought method of migration learning to the entrepreneurial borrowing scenario, provides a new solution to the cold start problem of the entrepreneurial borrowing risk control system and verifies the feasibility and effectiveness of the migration learning method applied in the risk control field through empirical data.

Medicine distribution logistics pattern in pharmaceutical supply chains is a hot topic, which aims to predict applicable and efficient medicine distribution patterns so that the medicine can be distributed effectively. This research aims to propose a new method, named density-distance method, that works based on Kth proximity using patient features (including age, gender, education, inherent diseases, systemic diseases and disorders); geographical features (city, state, population, density, land area) and supply chain features (destination and transportation system).

The proposed method of this research consists of two main phases where in the first phase, quantitative data analytics will be carried out to find out the significant factors and their impacts on medicine distribution. Then, in the next phase, a new Kth-proximity density-distance-based method is proposed to determine the best locations for the wholesalers while designing a supply chain.

The findings show that the proposed method can effectively design a supply chain network using realistic features. In addition, it is found that while the distance-density aggregate index is applied, the wholesalers' locations will be different compared to classic supply chain designs. The results show that age, public hygiene level and density are the most influential during designing new supply chains.

The outcomes of this research can be used in the medicine supply chains to predict appropriate medicine distribution logistics patterns.

In this research, the machine learning method based on the nearest neighbor has been used for the first time in the design of the supply chain network.

This paper proposes a self-supervised monocular depth estimation algorithm under multiple constraints, which can generate the corresponding depth map end-to-end based on RGB images. On this basis, based on the traditional visual simultaneous localisation and mapping (VSLAM) framework, a dynamic object detection framework based on deep learning is introduced, and dynamic objects in the scene are culled during mapping.

Typical SLAM algorithms or data sets assume a static environment and do not consider the potential consequences of accidentally adding dynamic objects to a 3D map. This shortcoming limits the applicability of VSLAM in many practical cases, such as long-term mapping. In light of the aforementioned considerations, this paper presents a self-supervised monocular depth estimation algorithm based on deep learning. Furthermore, this paper introduces the YOLOv5 dynamic detection framework into the traditional ORBSLAM2 algorithm for the purpose of removing dynamic objects.

Compared with Dyna-SLAM, the algorithm proposed in this paper reduces the error by about 13%, and compared with ORB-SLAM2 by about 54.9%. In addition, the algorithm in this paper can process a single frame of image at a speed of 15–20 FPS on GeForce RTX 2080s, far exceeding Dyna-SLAM in real-time performance.

This paper proposes a VSLAM algorithm that can be applied to dynamic environments. The algorithm consists of a self-supervised monocular depth estimation part under multiple constraints and the introduction of a dynamic object detection framework based on YOLOv5.

The main purpose of this analysis was to demonstrate the value of predictive modeling of student success and identify the key groups of students for which library instruction could provide the most impact.

Data regarding the attendance of library instruction associated with a first-year writing course were combined with student demographic and academic data over a four year period representing over 10,000 students. We applied supervised machine learning methods to determine the most accurate model for predicting student outcomes, including course outcome, persistence and graduation. We also assessed the impact of library instruction on these outcomes.

The gradient-boosted decision tree model provided the most accurate predictions. The impact of library instruction was modest but still was second only to the previous grade point average (GPA). The value of this metric, however, was greatest for students who were struggling, especially those who were first-generation students, regardless of ethnicity. More notably, the impact of library instruction was substantially greater for specific student demographics, including students with lower cumulative GPAs.

Features of the models were limited to high-level academic metrics, some of which may not be very useful in predicting outcomes. Measures more closely related to learning styles, the course or course of study could provide for greater accuracy.

Prediction modeling could allow for a more selective approach to outreach and offers information that the librarian can use to customize instruction sessions and reference interactions.

Targeting students who may be at risk of not succeeding in a course has ethical implications either way. If used to bias the subjective assessments, these predictions could produce self-fulfilling prophecies. Conversely, to ignore indicators of possible difficulties the student may have with the material is a disservice to the education of that student.

There are few studies that have incorporated library instruction into models of predicting student outcomes. Library resources and services can play a major role in the success of students, particularly those who have had less exposure to the resources and skills needed to use these resources.

Taking into consideration the current human need for agricultural produce such as rice that requires water for growth, the optimal consumption of this valuable liquid is important. Unfortunately, the traditional use of water by humans for agricultural purposes contradicts the concept of optimal consumption. Therefore, designing and implementing a mechanized irrigation system is of the highest importance. This system includes hardware equipment such as liquid altimeter sensors, valves and pumps which have a failure phenomenon as an integral part, causing faults in the system. Naturally, these faults occur at probable time intervals, and the probability function with exponential distribution is used to simulate this interval. Thus, before the implementation of such high-cost systems, its evaluation is essential during the design phase.

The proposed approach included two main steps: offline and online. The offline phase included the simulation of the studied system (i.e. the irrigation system of paddy fields) and the acquisition of a data set for training machine learning algorithms such as decision trees to detect, locate (classification) and evaluate faults. In the online phase, C5.0 decision trees trained in the offline phase were used on a stream of data generated by the system.

The proposed approach is a comprehensive online component-oriented method, which is a combination of supervised machine learning methods to investigate system faults. Each of these methods is considered a component determined by the dimensions and complexity of the case study (to discover, classify and evaluate fault tolerance). These components are placed together in the form of a process framework so that the appropriate method for each component is obtained based on comparison with other machine learning methods. As a result, depending on the conditions under study, the most efficient method is selected in the components. Before the system implementation phase, its reliability is checked by evaluating the predicted faults (in the system design phase). Therefore, this approach avoids the construction of a high-risk system. Compared to existing methods, the proposed approach is more comprehensive and has greater flexibility.

By expanding the dimensions of the problem, the model verification space grows exponentially using automata.

Unlike the existing methods that only examine one or two aspects of fault analysis such as fault detection, classification and fault-tolerance evaluation, this paper proposes a comprehensive process-oriented approach that investigates all three aspects of fault analysis concurrently.

Historical newspaper collections provide a wealth of information about the past. Although the digitization of these collections significantly improves their accessibility, a large portion of digitized historical newspaper collections, such as those of KBR, the Royal Library of Belgium, are not yet searchable at article-level. However, recent developments in AI-based research methods, such as document layout analysis, have the potential for further enriching the metadata to improve the searchability of these historical newspaper collections. This paper aims to discuss the aforementioned issue.

In this paper, the authors explore how existing computer vision and machine learning approaches can be used to improve access to digitized historical newspapers. To do this, the authors propose a workflow, using computer vision and machine learning approaches to (1) provide article-level access to digitized historical newspaper collections using document layout analysis, (2) extract specific types of articles (e.g. feuilletons – literary supplements from Le Peuple from 1938), (3) conduct image similarity analysis using (un)supervised classification methods and (4) perform named entity recognition (NER) to link the extracted information to open data.

The results show that the proposed workflow improves the accessibility and searchability of digitized historical newspapers, and also contributes to the building of corpora for digital humanities research. The AI-based methods enable automatic extraction of feuilletons, clustering of similar images and dynamic linking of related articles.

The proposed workflow enables automatic extraction of articles, including detection of a specific type of article, such as a feuilleton or literary supplement. This is particularly valuable for humanities researchers as it improves the searchability of these collections and enables corpora to be built around specific themes. Article-level access to, and improved searchability of, KBR's digitized newspapers are demonstrated through the online tool (https://tw06v072.ugent.be/kbr/).

The energy generation process through photovoltaic (PV) panels is contingent upon uncontrollable variables such as wind patterns, cloud cover, temperatures, solar irradiance intensity and duration of exposure. Fluctuations in these variables can lead to interruptions in power generation and losses in output. This study aims to establish a measurement setup that enables monitoring, tracking and prediction of the generated energy in a PV energy system to ensure overall system security and stability. Toward this goal, data pertaining to the PV energy system is measured and recorded in real-time independently of location. Subsequently, the recorded data is used for power prediction.

Data obtained from the experimental setup include voltage and current values of the PV panel, battery and load; temperature readings of the solar panel surface, environment and the battery; and measurements of humidity, pressure and radiation values in the panel’s environment. These data were monitored and recorded in real-time through a computer interface and mobile interface enabling remote access. For prediction purposes, machine learning methods, including the gradient boosting regressor (GBR), support vector machine (SVM) and k-nearest neighbors (k-NN) algorithms, have been selected. The resulting outputs have been interpreted through graphical representations. For the numerical interpretation of the obtained predictive data, performance measurement criteria such as mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE) and R-squared (R²) have been used.

It has been determined that the most successful prediction model is k-NN, whereas the prediction model with the lowest performance is SVM. According to the accuracy performance comparison conducted on the test data, k-NN exhibits the highest accuracy rate of 82%, whereas the accuracy rate for the GBR algorithm is 80%, and the accuracy rate for the SVM algorithm is 72%.

The experimental setup used in this study, including the measurement and monitoring apparatus, has been specifically designed for this research. The system is capable of remote monitoring both through a computer interface and a custom-developed mobile application. Measurements were conducted on the Karabük University campus, thereby revealing the energy potential of the Karabük province. This system serves as an exemplary study and can be deployed to any desired location for remote monitoring. Numerous methods and techniques exist for power prediction. In this study, contemporary machine learning techniques, which are pertinent to power prediction, have been used, and their performances are presented comparatively.

This study aims to compare and evaluate the application of commonly used machine learning (ML) algorithms used to develop models for assessing energy efficiency of buildings.

This study foremostly combined building energy efficiency ratings from several data sources and used them to create predictive models using a variety of ML methods. Secondly, to test the hypothesis of ensemble techniques, this study designed a hybrid stacking ensemble approach based on the best performing bagging and boosting ensemble methods generated from its predictive analytics.

Based on performance evaluation metrics scores, the extra trees model was shown to be the best predictive model. More importantly, this study demonstrated that the cumulative result of ensemble ML algorithms is usually always better in terms of predicted accuracy than a single method. Finally, it was discovered that stacking is a superior ensemble approach for analysing building energy efficiency than bagging and boosting.

While the proposed contemporary method of analysis is assumed to be applicable in assessing energy efficiency of buildings within the sector, the unique data transformation used in this study may not, as typical of any data driven model, be transferable to the data from other regions other than the UK.

This study aids in the initial selection of appropriate and high-performing ML algorithms for future analysis. This study also assists building managers, residents, government agencies and other stakeholders in better understanding contributing factors and making better decisions about building energy performance. Furthermore, this study will assist the general public in proactively identifying buildings with high energy demands, potentially lowering energy costs by promoting avoidance behaviour and assisting government agencies in making informed decisions about energy tariffs when this novel model is integrated into an energy monitoring system.

This study fills a gap in the lack of a reason for selecting appropriate ML algorithms for assessing building energy efficiency. More importantly, this study demonstrated that the cumulative result of ensemble ML algorithms is usually always better in terms of predicted accuracy than a single method.

This study aims to predict customer churn in resort hotels by calculating the churn probability of repeat customers for future stays in the same hotel brand.

Based on the recency, frequency, monetary (RFM) paradigm, random forest and logistic regression supervised machine learning algorithms were used to predict churn behavior. The model with superior performance was used to detect potential churners and generate a priority matrix.

The random forest algorithm showed a higher prediction performance with an 80% accuracy rate. The most important variables were RFM-based, followed by hotel sector-specific variables such as market, season, accompaniers and booker. Some managerial strategies were proposed to retain future churners, clustered as “hesitant,” “economy,” “alternative seeker,” and “opportunity chaser” customer groups.

This study contributes to the theoretical understanding of customer behavior in the hospitality industry and provides valuable insight for hotel practitioners by demonstrating the methods that facilitate the identification of potential churners and their characteristics.

Most customer retention studies in hospitality either concentrate on the antecedents of retention or customers’ revisit intentions using traditional methods. Taking a unique place within the literature, this study conducts churn prediction analysis for repeat hotel customers by opening a new area for inquiry in hospitality studies.