Random Forest Classification Algorithm Research Articles

Comparative Analysis Of Machine Learning And Deep Learning Models Integrated With Altman Z-Score For Financial Distress Prediction In Companies Listed On The Indonesia Stock Exchange (IDX)

Financial distress is a condition in which a company experiences a decline in its financial situation, which is typically temporary. However, financial distress can worsen if not promptly addressed, leading to bankruptcy. Early identification of potential financial distress in a company is crucial for stakeholders such as investors, creditors, and regulators. In practice, predicting financial distress in a company is not an easy task. One of the methods commonly used to identify early symptoms of financial distress is the method introduced by Altman in 1986. Altman's research model, known as the Z-Score, determines a value based on standard calculations of financial ratios to indicate the likelihood of a company experiencing bankruptcy. The use of artificial intelligence, such as deep learning, can enhance the scope of research on the analysis and prediction of financial distress. This study aims to conduct a comparative analysis of machine learning models, such as the K-Nearest Neighbor Classifier and Random Forest Classifier, and deep learning models, such as Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN). This analysis is carried out to evaluate the success rate of the proposed deep learning models in predicting financial distress in companies operating in Indonesia. Based on the conducted research, the K-Nearest Neighbor Classifier algorithm achieved an accuracy of 77.42% during testing and 88.89% during validation, the Random Forest Classifier algorithm achieved an accuracy of 87.09% during testing and 95.24% during validation, the CNN model achieved an accuracy of 95.16% during testing and 96.83% during validation, and the RNN model achieved an accuracy of 93.55% during testing and 96.83% during validation. Based on these results, the deep learning method has a higher average success rate than machine learning models in predicting financial distress.

Team Performance Indicators That Predict Match Outcome in Rugby Union

The aim of the study is to identify the most significant indicators of the national team's performance at the European Rugby Championships 15 and to design a model for predicting the outcomes of matches. Data was collected from teams’ performance at the European Rugby 15 Championships 2021, 2022 and 2023 for the analysis. The total number of matches was 41. All indicators presented in the official reports were taken: 22 for the home and away teams. The analysis of the team results was carried out according to all indicators: mean value, standard deviation, and test were used to compare the performance indicators of the winning and losing teams. Machine learning techniques were utilized to develop a predictive model for match outcomes. On one hand, 15 indicators (68.2%) are higher for teams that won (winning teams). On the other hand, 7 (31.8%) indicators are higher for teams that lost. The difference between the teams' means varies from -56.46% (the minus indicates that this indicator is higher for the teams that lost) to 273.68%. Based on the results, the Random Forest Classifier and Extra Trees Classifier algorithms have the best prediction accuracy (0.92). The most significant indicators of team performance that affect the final result of the match are tries (196.3% – the difference between the average values of winning and losing teams), conversions (176.7%), missed tackles (- 56.46%), offload (126.3%). Based on the data obtained, refining the team training process in Rugby 15 is possible.

Player Winning Probability Model (PWPM): A Review

The Olympic Games are an international event and a source of pride for all countries. With over 200 countries participating in the Olympic Games, it is an international sporting event. The Olympic Games stand as a pinnacle of athletic achievement and global unity, showcasing the finest talents from diverse nations across numerous sporting disciplines. In this review we used machine Learning Algorithms to do prediction on Olympic medal and applying various tools and techniques for Data Analysis using historical data. The process involves Data Collection, Data Cleaning, Data Processing, Exploratory Data Analysis and Medal Prediction. The main objective of this work is to study the complete set of Olympic data to discover pattern and relationship between variables using analysis of data to assess how the Olympic Games changed over time and to do prediction on medals. In this work a PWP model has been proposed in which the Random Forest classifier algorithm has been selected for the classification of player winning probability with accuracy of 70.223325%.

A Very High-Resolution Urban Green Space from the Fusion of Microsatellite, SAR, and MSI Images

Jakarta holds the distinction of being the largest capital city among ASEAN countries and ranks as the second-largest metropolitan area in the world, following Tokyo. Despite numerous studies examining the diverse urban land use and land cover patterns within the city, the recent state of urban green spaces has not been adequately assessed and mapped precisely. Most previous studies have primarily focused on urban built-up areas and manmade structures. In this research, the first-ever detailed map of Jakarta’s urban green spaces as of 2023 was generated, with a resolution of three meters. This study employed a combination of supervised classification and evaluated two machine learning algorithms to achieve the highest accuracy possible. To achieve this, various satellite images were utilized, including VV and VH polarizations from Sentinel-1, multiple bands from Sentinel-2, and eight bands from Planet. The Planet data were subsequently transformed into the Red-Edge Triangulated Vegetation Index and Red-Edge Triangulated Wetness Index. The data training and testing samples for urban green spaces were obtained using the Street View images available on Google Maps. The results revealed that using the Random Forest classifier algorithm and only eight bands of Planet images achieved an accuracy rate of 84.9%, while a combination of multiple images achieved an impressive 95.9% accuracy rate. Jakarta’s urban areas cover approximately 33.2% of green spaces. This study provides unprecedented insights into the type, size, and spatial distribution of Jakarta’s urban green spaces, enabling urban residents and stakeholders to explore and promote healthier living and better manage these green areas. Additionally, a previously unexplored concept, Jakarta’s urban green belt, is introduced.

Machine Learning Techniques for Gender Recognition and Age Prediction Using Digital Images of Human Dentition

In today's digital age, the utilization of digital medical imagery is on the rise, since the increased computational power that enables the development of more advanced machine learning techniques. We have implemented a system capable of determining the gender and estimating the age of individuals based on digital images of their teeth. Teeth, being a robust and relatively unchanging part of the human body, serve as an ideal source for such analysis. By examining Orthopantomogram images, our system systematically extracts gender and age-related information. Developed using the Python programming language, our application is particularly valuable in forensic contexts, where traditional prediction methods can be time-consuming, often taking days. In contrast, our model delivers results in under a minute. We employed the Random Forest (RF) Classifier algorithm to deploy the prediction model, as it effectively addresses the overfitting issue associated with growing input sample sizes. With this approach, we have achieved an impressive accuracy rate of 94% in gender and age prediction.

A comprehensive analysis and future projection of land use and land cover dynamics in a fast-growing city: A case study of Sekondi-Takoradi metropolis, Ghana

Sekondi-Takoradi is the most urbanized metropolis in the Western region of Ghana. Over the past two decades, it has experienced rapid shifts in both population density and land cover. Unfortunately, these transformations have significantly altered the natural landscape and expanded urban coverage. Previous studies have not established enough evidence on land use and land cover changes in the metropolis. For this reason, this study examined the existence, rate, and spatial distribution of land use and land cover as well as the future changes in land cover in the metropolis. Landsat imagery spanning from 1991, 2009, 2016, to 2023 provided the basis for evaluating the spatiotemporal dynamics of land use and cover within the metropolis. The Random Forest Classification algorithm was employed to categorize the images into three distinct classes: water, vegetation, and built-up areas. Analysis revealed a rapid growth in built-up areas by 63.08 km² (32.91 %), accompanied by a decrease in vegetation and water coverage by 60.99 km² (31.82 %) and 2.08 km² (1.09 %) respectively. This highlights the accelerating urbanization trend, emphasizing the critical need for vigilant monitoring and controlled urban expansion to mitigate potential adverse effects. Projections for 2030 indicate further changes with water areas decreasing to 1.21 km² (0.63 %), vegetation diminishing to 95.31 km² (49.73 %), and built-up areas expanding to 95.14 km² (49.64 %). This signifies a significant shift towards built-up areas, altering the metropolis' land cover composition and potentially impeding the achievement of Sustainable Development Goal 11. Understanding these changes in land use and land cover holds significant value for policymakers, enabling them to effectively monitor and safeguard natural resources.

Rapid Production Method of Massive Thematic Maps Based on Geospatial Knowledge Extraction

<p>Geospatial knowledge in massive academic papers can provide knowledge services such as location-based research hotspot analysis, spatio-temporal data aggregation, research results recommendation, etc. However, geospatial knowledge often exists implicitly in literature resources in unstructured form, which is difficult to be directly accessed and mined and utilized for rapid production of massive thematic maps. In this paper, we take the geospatial knowledge of the area studied as an example and introduce its extraction method in detail. An integrated feature template matching and random forest classification algorithm is proposed for accurately identifying research areas from the abstract texts of academic papers and producing thematic maps. Firstly, the precise recognition of geographical names is achieved step by step based on BiLSTM-CRF algorithm and improved heuristic disambiguation method; then, the area studied is extracted by the designed integrated feature recognition template of area studied using random forest classification algorithm, and a fast thematic map is designed for the knowledge of area studied, topic and literature. The experimental results show that the area studied recognition accuracy can reach 97%, the F-value is 96%, and the recall rate reaches 96%, achieving high accuracy and high efficiency of area studied extraction in text. Based on the geospatial knowledge, the thematic map can achieve the effect of fast map formation and accurate expression.</p> <p> </p>

Land Use Land Cover (LULC) Change Dynamics Associated with Mining Activities in Kitwe District and Adequacy of the Legal Framework on Mine Closure in Zambia

Land use land cover (LULC) changes resulting from copper exploration in Kitwe District, Copperbelt Province has adversely impacted the environment. To understand LULC change dynamics associated with mining activities, this study mapped LULC changes using the Google Earth Engine (GEE) from 1990 to 2020. In addition, the Zambian legal framework for mine closure was assessed in terms of adequacy and comprehensiveness. A remote sensing analysis using Landsat TM (1990, 2000, and 2010) and OLI (2020) images was performed and the GEE Random Forest classifier algorithm was employed to detect LULC changes. Then, transition matrices and overall changes were calculated for each LULC class. The LULC classification had an overall accuracy and kappa coefficient of 82.47% and 0.78, respectively. In total, 45.2% of the district area (360.92 km2) experienced LULC changes from 1990 to 2020. The overall change indicates that the areas of built-up area, bare land, and grassland/pasture/agricultural land gained 35.84, 14.67, and 43.53 km2, respectively, while forest lost 95.30 km2, with the major driver being the privatization of mining companies. Several concerns regarding the mine closure process practiced in Zambia have principally been raised to the government. Although the legislation generally conformed to international best practices, a gap involving various pieces of legislation, overlapping requirements, and different interpretations of the laws by different governmental departments makes the system complex and unmanageable. An area of concern is the government’s capability and competence to implement legislation. Ineffective law enforcement, that is, the inadequacy of the legislation, is to blame for LULC changes in mining areas, resulting in mining corporations not paying attention to the changes made, particularly regarding mine closures. This study provides decision-makers and land use planners with baseline knowledge on LULC changes that can be valuable for future mining legislation and how these legislations can be effectively executed to ensure sustainable mine closure.

Spatio-Temporal Changes and Driving Forces Analysis of Urban Open Spaces in Shanghai between 1980 and 2020: An Integrated Geospatial Approach

Although there is extensive research demonstrating the significant loss and fragmentation of urban spaces caused by rapid urbanization, to date, no empirical research in Shanghai has investigated the spatiotemporal dynamics of urban open spaces using a comprehensive set of integrated geospatial techniques based on long-sequence time series. Based on the Google Earth Engine (GEE) platform and using the Random Forest (RF) classifier, multiple techniques, namely landscape metrics, trend analysis, open space ratio, transition matrix, Normalized Difference Vegetation Index (NDVI), and fractal dimension analysis, were applied to analyze the Landsat satellite data. Next, Geographic Detector (GeoDetector) methods were used to investigate the driving forces of such spatial variations. The results showed that (1) the RF classification algorithm, supported by the GEE, can accurately and quickly obtain a research object dataset, and that calculating the optimal spatial grain size for open space pattern was 70 m; (2) open spaces exhibited declining and contracting trends; and open spaces in the city experienced a decline from 91.83% in 1980 to 69.63% in 2020. Meanwhile, the degree of open spaces in each district increased to different extents, whilst connectivity markedly decreased. Furthermore, the open space of city center districts showed the lowest rate of decrease, with open space patterns fragmenting due to encroaching urbanization; (3) the contribution of socioeconomic factors to the spatial–temporal changes in open space continually has increased over the past 40 years, and were also higher than natural geographic factors to some extent. Apart from offering policy insights guiding the future spatial planning and development of the city, this paper has contributions from both methodological and empirical perspectives. Based on integrated remote sensing and geographic information science (GIS) techniques, this paper provides updated evidence and a clearer understanding of the spatiotemporal variations in urban spaces and their influencing mechanisms in Shanghai.

Heart Disease Detection Using AI

Over the past few decades, cardiovascular disease has emerged as the primary cause of death worldwide in both industrialized and developing nations. Early detection of heart problems and continued clinical monitoring can reduce death rates. However, because it takes more time and experience, it is not possible to accurately detect heart disorders in all cases and to have a specialist talk with a patient for 24 hours. We demonstrate how machine learning can be used to estimate an individual's risk of developing heart disease. This study presents data processing, which includes converting categorical columns and working with categorical variables. We outline the three primary stages of developing an application: gathering datasets, running logistic regression, and assessing the properties of the dataset. The random forest classifier technique is developed to diagnose cardiac problems more precisely. Data analysis is needed for this application since it is considered noteworthy. The random forest classifier algorithm, which improves the accuracy of research diagnosis, is next covered, along with the experiments and findings.

Disposal of solid waste from civil construction: a screening proposal for a suitability system and case study in Nepomuceno, Minas Gerais

ABSTRACT Most Brazilian municipalities do not have regulated areas for solid waste disposal in civil construction. Usually, residues are disposed of vacant lots and dumps, posing risks to the population health and the environment. Soils are the primary means for the disposal or recycling of waste, highlighting the importance of well-characterized soils and their respective landscape. This study aimed to establish a land suitability system for solid residues in civil construction and apply such information in a case study in Southeastern Brazil. An unprecedented digital soil map with a resolution of 30 m was created using the random forest classifier algorithm and soil field prospection information. A guide listing favorable soil-landscape attributes that most prevent soil erosion, water bodies or water table contamination was elaborated and discussed. Thus, such information was linked through a suitability system to classify areas with potential for receiving waste on a daily volume basis as follows: large size: >500 m³ day -1 , medium size: >100 m³ day -1 and <300 m³ day -1 , and small size <100 m³ day -1 . Topography and soil depth were the most limiting factors of the areas in the case study. The proposed attributes as criteria for the suitability system complement the current state legislation. A total of 236 ha closer to the urban perimeter connected by roads in good condition were classified as suitable for managing medium- and small-scale daily volume, whose destination might reduce transportation and installation costs in the study area.

Gut microbiota predicts the diagnosis of ulcerative colitis in Saudi children.

Ulcerative colitis (UC) is an immune-mediated chronic inflammatory condition with a worldwide distribution. Although the etiology of this disease is still unknown, the understanding of the role of the microbiota is becoming increasingly strong. To investigate the predictive power of the gut microbiota for the diagnosis of UC in a cohort of newly diagnosed treatment-naïve Saudi children with UC. The study population included 20 children with a confirmed diagnosis of UC and 20 healthy controls. Microbial DNA was extracted and sequenced, and shotgun metagenomic analysis was performed for bacteria and bacteriophages. Biostatistics and bioinformatics demonstrated significant dysbiosis in the form of reduced alpha diversity, beta diversity, and significant difference of abundance of taxa between children with UC and control groups. The receiver operating characteristic curve, a probability curve, was used to determine the difference between the UC and control groups. The area under the curve (AUC) represents the degree of separability between the UC group and the control group. The AUC was calculated for all identified bacterial species and for bacterial species identified by the random forest classification algorithm as important potential biomarkers of UC. A similar method of AUC calculation for all bacteriophages and important species was used. The median age and range were 14 (0.5-21) and 12.9 (6.8-16.3) years for children with UC and controls, respectively, and 40% and 35% were male for children with UC and controls, respectively. The AUC for all identified bacterial species was 89.5%. However, when using the bacterial species identified as important by random forest classification algorithm analysis, the accuracy increased to 97.6%. Similarly, the AUC for all the identified bacteriophages was 87.4%, but this value increased to 94.5% when the important bacteriophage biomarkers were used. The very high to excellent AUCs of fecal bacterial and viral species suggest the potential use of noninvasive microbiota-based tests for the diagnosis of unusual cases of UC in children. In addition, the identification of important bacteria and bacteriophages whose abundance is reduced in children with UC suggests the potential of preventive and adjuvant microbial therapy for UC.

Smart Urban Cadastral Map Enrichment—A Machine Learning Method

Enriching and updating maps are among the most important tasks of any urban management organization for informed decision making. Urban cadastral map enrichment is a time-consuming and costly process, which needs an expert’s opinion for quality control. This research proposes a smart framework to enrich a cadastral base map using a more up-to-date map automatically by machine learning algorithms. The proposed framework has three main steps, including parcel matching, parcel change detection and base map enrichment. The matching step is performed by checking the center point of each parcel in the other map parcels. Support vector machine and random forest classification algorithms are used to detect the changed parcels in the base map. The proposed models employ the genetic algorithm for feature selection and grey wolf optimization and Harris hawks optimization for hyperparameter optimization to improve accuracy and performance. By assessing the accuracies of the models, the random forest model with feature selection and grey wolf optimization, with an F1-score of 0.9018, was selected for the parcel change detection method. Finally, the detected changed parcels in the base map are deleted and relocated automatically with corresponding parcels in the more up-to-date map by the affine transformation.

Data-driven casting defect prediction model for sand casting based on random forest classification algorithm

Data-driven casting defect prediction model for sand casting based on random forest classification algorithm

Analysis of Credit Card Fraud Detection Performance Using Random Forest Classifier & Neural Networks Model

Credit card fraud is one example of data manipulation in the e-commerce industry. Because credit card fraud is so common, preventing it can be difficult. Therefore, it is important to identify credit card fraud as soon as it occurs. Determining the validity of a transaction is a fraud detection process. Credit/ Debit cards/ any financial services are small plastic cards given to members of certain financial organizations with proper identity and verification. This research is quantitative research that uses a dataset that has been verified and classified as fraudulent or non-fraudulent transactions. This research will limit itself to credit card fraud detection methods that use Neural Networks and Random Forest Classifier algorithms. Using publicly available credit card transaction data as a starting point, this study will examine how well machine learning algorithms do overall in identifying credit card fraud. First of all, we check for the presence of duplicate values in the data set. The result of this check shows that there are no duplicate values in the set, marked with the value ‘False’. Next, we focus on class in the target variable ‘Class’. The output of this check shows the wide variety of samples for each class in the goal variable ‘Class’. In this example, class 0 has 284.315 samples and class 1 has 284.315 samples. Based totally on the studies performed, it is possible to conclude that the usage of machine learning algorithms, together with Random Forest Classifier and Neural Networks, can be effective in detecting credit card fraud. The results of the algorithm performance analysis show that the model developed can identify fraudulent transactions accurately. In addition, the data preparation stages, correlation analysis, and model performance evaluation also contribute to understanding potentially fraudulent credit card transaction patterns.

An Efficient Machine Learning Prediction Method for Vehicle Detection: Data Analytics Framework

The availability of transportation is considered a significant hallmark of a developed society. Since the evolution of the human species, the imperative to relocate from one location to another has been a fundamental requirement. At present, there exists a plethora of transportation options in Indonesia. However, most individuals favor road transportation due to its ease and convenience. The rise in population has led to a corresponding increase in the number of vehicles on the roadways. Hence, it presents a challenge for security authorities and governmental bodies to oversee all automobiles' mobility across various locations effectively. The present study proposes a methodology for detecting and tracking vehicles using video-based techniques. The process's initial stages involve preprocessing, including frame conversion and background subtraction. Next, the process of detecting vehicles involves the utilization of change detection and a model of body shape. Subsequently, the next stage entails the feature extraction process, focusing on extracting energy features and directional cosine. Subsequently, a technique for optimizing data is employed on the vector comprising excessively extracted features. The methodology integrates a data mining technique based on association rules, which is subsequently complemented by a random forest classification algorithm. The approach generally integrates multiple methodologies to attain effective and precise identification of automobiles in video-derived datasets.

Designing A Pdf Malware Detection System Using Machine Learning

This research proposes an approach to build malicious PDF detection system using random forest algorithm, focusing the Evasive-PDFMal2022 dataset which is updated and extended with the addition of new datasets. This dataset includes malicious PDF files from CVE and Exploit-DB, non-malicious PDF files, as well as files from private collections and Technically-oriented PDF Collection. Features were extracted using the PDFID tool, resulting in 29 structural features that formed the basis for the Random Forest classification algorithm. Experiments showed that the model trained with the new dataset provided accuracy equivalent to the Evasive-PDFMal2022 model, at 98%, albeit with a small decrease in recall for the benign class. In addition, this research involved the creation of a website for metadata extraction and malicious PDF detection. Recognition goes to the dataset contributors, tool developers, and dataset providers from NIST and Exploit-DB. Overall, this research successfully increased the representation and diversity of the dataset, provided good model training results, improved detection from 3 malicious PDF variants to 13 variants, and created a practical tool for malicious PDF extraction and detection. Nonetheless, further development may be required to improve detection performance in more complex scenarios

Random Forest Machine Learning for Spam Email Classification

This research discusses the crucial role of email as a main element in digital communication, facilitating information transfer and serving as an advertising platform. However, the problem of email spam, which involves sending unsolicited commercial messages, has had negative impacts such as consuming large amounts of resources and disrupting user experience. With its affordable cost and ease of sending messages to thousands of recipients, email spam includes product promotions, pornographic material, viruses and irrelevant content. The impact includes loss of time and damage to the user's computer resources. To address this problem, email services provide advanced spam filters that use email content analysis and machine learning techniques. This research focuses on the use of the Random Forest Classification algorithm as a basis for filtering spam emails. Although Random Forest is known to have strong classification capabilities, the risk of overfitting is a challenge. Therefore, this study adopts the Randomized Search CV method to identify the best parameter combination, ensuring the reliability of the model in dealing with the complexity of diverse email datasets. With this approach, this research contributes to the development of effective solutions to reduce the impact of email spam in digital communications.

Predicting Contralateral Surgery for Trapeziometacarpal Arthrosis Within 5 Years

Symptomatic trapeziometacarpal (TMC) joint arthritis is a common cause of hand pain. It is unknown how many patients ultimately elect to have bilateral surgery for TMC arthritis. In this study, we assessed the frequency and predictive factors for contralateral TMC surgery in patients who underwent prior TMC surgery. We identified 712 patients who underwent primary surgery for TMC arthritis with a follow-up period of 5 years. We collected demographic, surgical, and follow-up data. Prediction models for contralateral surgery using a training and testing data set were created with multivariable logistic regression and random forest classifier algorithms. At the time of initial surgery, 230 patients had bilateral thumb pain (32%), but only 153 patients ultimately had an operation for TMC arthritis on the contralateral side within 5 years (21% of 712 total patients and 67% of 230 patients with bilateral pain). Common predictive factors between both models for contralateral surgery were younger age (odds ratio [OR]= 0.95; 95% confidence interval [CI], 0.93-0.98), bilateral thumb pain (OR= 3.76; 95% CI, 2.52-5.65), and anxiety disorders (OR= 1.84; 95% CI, 1.11-3.03). In our study, we found that the rate of contralateral surgery was 21% in patients who underwent prior TMC surgery. Predictive factors for future contralateral surgery included younger age, bilateral thumb pain, and anxiety disorder at the time of initial surgery. Prognostic II.

Random forest analysis of midbrain hypometabolism using [18F]-FDG PET identifies Parkinson's disease at the subject-level.

Parkinson's disease (PD) is currently diagnosed largely on the basis of expert judgement with neuroimaging serving only as a supportive tool. In a recent study, we identified a hypometabolic midbrain cluster, which includes parts of the substantia nigra, as the best differentiating metabolic feature for PD-patients based on group comparison of [18F]-fluorodeoxyglucose ([18F]-FDG) PET scans. Longitudinal analyses confirmed progressive metabolic changes in this region and, an independent study showed great potential of nigral metabolism for diagnostic workup of parkinsonian syndromes. In this study, we applied a machine learning approach to evaluate midbrain metabolism measured by [18F]-FDG PET as a diagnostic marker for PD. In total, 51 mid-stage PD-patients and 16 healthy control subjects underwent high-resolution [18F]-FDG PET. Normalized tracer update values of the midbrain cluster identified by between-group comparison were extracted voxel-wise from individuals' scans. Extracted uptake values were subjected to a random forest feature classification algorithm. An adapted leave-one-out cross validation approach was applied for testing robustness of the model for differentiating between patients and controls. Performance of the model across all runs was evaluated by calculating sensitivity, specificity and model accuracy for the validation data set and the percentage of correctly categorized subjects for test data sets. The random forest feature classification of voxel-based uptake values from the midbrain cluster identified patients in the validation data set with an average sensitivity of 0.91 (Min: 0.82, Max: 0.94). For all 67 runs, in which each of the individuals was treated once as test data set, the test data set was correctly categorized by our model. The applied feature importance extraction consistently identified a subset of voxels within the midbrain cluster with highest importance across all runs which spatially converged with the left substantia nigra. Our data suggest midbrain metabolism measured by [18F]-FDG PET as a promising diagnostic imaging tool for PD. Given its close relationship to PD pathophysiology and very high discriminatory accuracy, this approach could help to objectify PD diagnosis and enable more accurate classification in relation to clinical trials, which could also be applicable to patients with prodromal disease.