Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
3.5
2.4
Journals
Mathematics
Special Issues
Multidisciplinary Models and Applications of Machine Learning and Computational Statistics
share announcement

Multidisciplinary Models and Applications of Machine Learning and Computational Statistics

A special issue of Mathematics (ISSN 2227-7390). This special issue belongs to the section "Mathematics and Computer Science".

Deadline for manuscript submissions: closed (25 March 2023) | Viewed by 14309

Special Issue Editor

Dr. Dan Vilenchik

E-Mail Website
Guest Editor
School of Electrical and Computer Engineering, Ben-Gurion University of the Negev, Beer-Sheva 84105, Israel
Interests: XAI for deep learning; concept learning; NLP; high-dimensional data; computational statistics; probabilistic methods in combinatorics; statistical inference
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

Machine learning (ML) algorithms have been applied successfully in many disciplines, taking on various types of data that were not accessible before, including big-data and high-dimensional data.
In this special issue we are interested in multidisciplinary research of machine learning and other disciplines, such as biology, chemistry, medicine and psychology.
Specifically, we are interested in case-studies that demonstrate how the ML methodology is improved by the added value that a domain expert brings to the research team.
A domain expert may contribute their special knowledge in key junctions of the data mining process:
data preprocessing, feature engineering, choosing the right ML model, tweaking the model in a non-standard way, and interpretation. One such example is ML-assisted drug design, where the knowledge of the chemist is crucial to extracting relevant features from the molecule.

Dr. Dan Vilenchik
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Mathematics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • machine learning (ML)
  • multidisciplinary research
  • domain-expert enhanced ML
  • interpretability
  • applications

Published Papers (6 papers)

Download All Papers
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:

Research

24 pages, 6989 KiB  
Article
LDA-CNN: Linear Discriminant Analysis Convolution Neural Network for Periocular Recognition in the Wild
by Amani Alahmadi, Muhammad Hussain and Hatim Aboalsamh
Mathematics 2022, 10(23), 4604; https://0-doi-org.brum.beds.ac.uk/10.3390/math10234604 - 05 Dec 2022
Cited by 1 | Viewed by 2630
Abstract
Due to the COVID-19 pandemic, the necessity for a contactless biometric system able to recognize masked faces drew attention to the periocular region as a valuable biometric trait. However, periocular recognition remains challenging for deployments in the wild or in unconstrained environments where [...] Read more.
Due to the COVID-19 pandemic, the necessity for a contactless biometric system able to recognize masked faces drew attention to the periocular region as a valuable biometric trait. However, periocular recognition remains challenging for deployments in the wild or in unconstrained environments where images are captured under non-ideal conditions with large variations in illumination, occlusion, pose, and resolution. These variations increase within-class variability and between-class similarity, which degrades the discriminative power of the features extracted from the periocular trait. Despite the remarkable success of convolutional neural network (CNN) training, CNN requires a huge volume of data, which is not available for periocular recognition. In addition, the focus is on reducing the loss between the actual class and the predicted class but not on learning the discriminative features. To address these problems, in this paper we used a pre-trained CNN model as a backbone and introduced an effective deep CNN periocular recognition model, called linear discriminant analysis CNN (LDA-CNN), where an LDA layer was incorporated after the last convolution layer of the backbone model. The LDA layer enforced the model to learn features so that the within-class variation was small, and the between-class separation was large. Finally, a new fully connected (FC) layer with softmax activation was added after the LDA layer, and it was fine-tuned in an end-to-end manner. Our proposed model was extensively evaluated using the following four benchmark unconstrained periocular datasets: UFPR, UBIRIS.v2, VISOB, and UBIPr. The experimental results indicated that LDA-CNN outperformed the state-of-the-art methods for periocular recognition in unconstrained environments. To interpret the performance, we visualized the discriminative power of the features extracted from different layers of the LDA-CNN model using the t-distributed Stochastic Neighboring Embedding (t-SNE) visualization technique. Moreover, we conducted cross-condition experiments (cross-light, cross-sensor, cross-eye, cross-pose, and cross-database) that proved the ability of the proposed model to generalize well to different unconstrained conditions. Full article
(This article belongs to the Special Issue Multidisciplinary Models and Applications of Machine Learning and Computational Statistics)
►▼ Show Figures

Figure 1

23 pages, 1561 KiB  
Article
One Aggregated Approach in Multidisciplinary Based Modeling to Predict Further Students’ Education
by Milan Ranđelović, Aleksandar Aleksić, Radovan Radovanović, Vladica Stojanović, Milan Čabarkapa and Dragan Ranđelović
Mathematics 2022, 10(14), 2381; https://0-doi-org.brum.beds.ac.uk/10.3390/math10142381 - 06 Jul 2022
Cited by 2 | Viewed by 1303
Abstract
In this paper, one multidisciplinary-applicable aggregated model has been proposed and verified. This model uses traditional techniques, on the one hand, and algorithms of machine learning as modern techniques, on the other hand, throughout the determination process of the relevance of model attributes [...] Read more.
In this paper, one multidisciplinary-applicable aggregated model has been proposed and verified. This model uses traditional techniques, on the one hand, and algorithms of machine learning as modern techniques, on the other hand, throughout the determination process of the relevance of model attributes for solving any problems of multicriteria decision. The main goal of this model is to take advantage of both approaches and lead to better results than when the techniques are used alone. In addition, the proposed model uses feature selection methodology to reduce the number of attributes, thus increasing the accuracy of the model. We have used the traditional method of regression analysis combined with the well-known mathematical method Analytic Hierarchy Process (AHP). This approach has been combined with the application of the ReliefF classificatory modern ranking method of machine learning. Last but not least, the decision tree classifier J48 has been used for aggregation purposes. Information on grades of the first-year graduate students at the Criminalistics and Police University, Belgrade, after they chose and finished one of the three possible study modules, was used for the evaluation of the proposed model. To the best knowledge of the authors, this work is the first work when considering mining closed frequent trees in case of the streaming of time-varying data. Full article
(This article belongs to the Special Issue Multidisciplinary Models and Applications of Machine Learning and Computational Statistics)
►▼ Show Figures

Figure 1

18 pages, 2105 KiB  
Article
Predicting Change in Emotion through Ordinal Patterns and Simple Symbolic Expressions
by Yair Neuman and Yochai Cohen
Mathematics 2022, 10(13), 2253; https://0-doi-org.brum.beds.ac.uk/10.3390/math10132253 - 27 Jun 2022
Cited by 4 | Viewed by 1409
Abstract
Human interlocutors may use emotions as an important signaling device for coordinating an interaction. In this context, predicting a significant change in a speaker’s emotion may be important for regulating the interaction. Given the nonlinear and noisy nature of human conversations and relatively [...] Read more.
Human interlocutors may use emotions as an important signaling device for coordinating an interaction. In this context, predicting a significant change in a speaker’s emotion may be important for regulating the interaction. Given the nonlinear and noisy nature of human conversations and relatively short time series they produce, such a predictive model is an open challenge, both for modeling human behavior and in engineering artificial intelligence systems for predicting change. In this paper, we present simple and theoretically grounded models for predicting the direction of change in emotion during conversation. We tested our approach on textual data from several massive conversations corpora and two different cultures: Chinese (Mandarin) and American (English). The results converge in suggesting that change in emotion may be successfully predicted, even with regard to very short, nonlinear, and noisy interactions. Full article
(This article belongs to the Special Issue Multidisciplinary Models and Applications of Machine Learning and Computational Statistics)
►▼ Show Figures

Figure 1

20 pages, 4793 KiB  
Article
Random Forest Winter Wheat Extraction Algorithm Based on Spatial Features of Neighborhood Samples
by Nayi Wang, Xiangsuo Fan, Jinlong Fan and Chuan Yan
Mathematics 2022, 10(13), 2206; https://0-doi-org.brum.beds.ac.uk/10.3390/math10132206 - 24 Jun 2022
Cited by 2 | Viewed by 3311
Abstract
In order to effectively obtain the winter wheat growing area in a large part of the Guanzhong plain, this paper proposes a random forest Guanzhong plain winter wheat extraction algorithm based on spatial features of neighborhood samples using the 250 m resolution spectral [...] Read more.
In order to effectively obtain the winter wheat growing area in a large part of the Guanzhong plain, this paper proposes a random forest Guanzhong plain winter wheat extraction algorithm based on spatial features of neighborhood samples using the 250 m resolution spectral imager (MERSI) of the FY-3 satellite as the data source. In this paper, first, the training and validation samples were obtained by constructing a neighborhood sample space sampling model, then the study area was classified using an integrated learning random forest Classifier, and finally the classification data obtained from different time phases were fused using voting game theory to obtain the final classification result map. The land use change and winter wheat distribution change from 2011 to 2014 were also analyzed. The experimental results showed that the overall accuracy of winter wheat obtained after random forest fusion processing was the highest compared with the traditional algorithm, reaching 98.63%. At the same time, LANDSAT 8 images were used to obtain the distribution of winter wheat, and the distribution areas obtained from MERSI data and LANDSAT 8 images were generally consistent in terms of spatial distribution as shown by the distribution areas at the county scale. Full article
(This article belongs to the Special Issue Multidisciplinary Models and Applications of Machine Learning and Computational Statistics)
►▼ Show Figures

Figure 1

20 pages, 9189 KiB  
Article
Who Is the Most Effective Country in Anti-Corruption? From the Perspective of Open Government Data and Gross Domestic Product
by Po-Yuan Shih, Cheng-Ping Cheng, Dong-Her Shih, Ting-Wei Wu and David C. Yen
Mathematics 2022, 10(13), 2180; https://0-doi-org.brum.beds.ac.uk/10.3390/math10132180 - 22 Jun 2022
Cited by 1 | Viewed by 2287
Abstract
Corruption represents the misuse of public power by government departments for personal gain, hindering a country’s economic growth. Corruption cannot be eliminated by implementing the national democratic system, and mature democratic countries also exist with varying degrees of corruption. Corruption affects people’s trust [...] Read more.
Corruption represents the misuse of public power by government departments for personal gain, hindering a country’s economic growth. Corruption cannot be eliminated by implementing the national democratic system, and mature democratic countries also exist with varying degrees of corruption. Corruption affects people’s trust in the public sector and the country’s economic development. Open government data can help people understand the governance performance of the government to reduce corruption in the public sector. Citizens can use open government data to generate innovative applications and economic value. This study uses a two-stage data envelopment analysis method to assess the anti-corruption efficiency of 21 countries from 2013 to 2017 through open government data, the corruption perception index, and GDP data. Then, the efficiency analyzed is introduced into the BCG (Boston Consulting Group) matrix to observe the distribution of these 21 countries. Analyzing the results showed that Uruguay and Costa Rica in Central and South America are the two most influential countries in fighting corruption. Turkey is at the bottom in the evaluation of anti-corruption efficiency. In addition, discussions of the included countries for their possible improvement in anti-corruption are also provided by using the association rule’s analysis. The study results will provide a reference for governments to effectively carry out anti-corruption work in the future. Full article
(This article belongs to the Special Issue Multidisciplinary Models and Applications of Machine Learning and Computational Statistics)
►▼ Show Figures

Figure 1

11 pages, 276 KiB  
Article
An Oblivious Approach to Machine Translation Quality Estimation
by Itamar Elmakias and Dan Vilenchik
Mathematics 2021, 9(17), 2090; https://0-doi-org.brum.beds.ac.uk/10.3390/math9172090 - 29 Aug 2021
Cited by 5 | Viewed by 1905
Abstract
Machine translation (MT) is being used by millions of people daily, and therefore evaluating the quality of such systems is an important task. While human expert evaluation of MT output remains the most accurate method, it is not scalable by any means. Automatic [...] Read more.
Machine translation (MT) is being used by millions of people daily, and therefore evaluating the quality of such systems is an important task. While human expert evaluation of MT output remains the most accurate method, it is not scalable by any means. Automatic procedures that perform the task of Machine Translation Quality Estimation (MT-QE) are typically trained on a large corpus of source–target sentence pairs, which are labeled with human judgment scores. Furthermore, the test set is typically drawn from the same distribution as the train. However, recently, interest in low-resource and unsupervised MT-QE has gained momentum. In this paper, we define and study a further restriction of the unsupervised MT-QE setting that we call oblivious MT-QE. Besides having no access no human judgment scores, the algorithm has no access to the test text’s distribution. We propose an oblivious MT-QE system based on a new notion of sentence cohesiveness that we introduce. We tested our system on standard competition datasets for various language pairs. In all cases, the performance of our system was comparable to the performance of the non-oblivious baseline system provided by the competition organizers. Our results suggest that reasonable MT-QE can be carried out even in the restrictive oblivious setting. Full article
(This article belongs to the Special Issue Multidisciplinary Models and Applications of Machine Learning and Computational Statistics)
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-6
Back to TopTop