Bayes Machine Learning Algorithm Research Articles

Weighted naïve bayes multi-user classification for adaptive authentication

Abstract Machine learning classification algorithms have been extensively utilized in addressing user authentication challenges. Nonetheless, a majority of solutions categorize users into three classes, whereas adaptive authentication scenarios necessitate classification beyond this threshold. The rationale behind this limitation has not been thoroughly explored. The current study leveraged the Naive Bayes theorem for user authentication endeavors to assess the risk associated with login attempts. The Naive Bayes Machine Learning algorithm, along with its variations such as Gaussian, Categorical, and Bernoulli, was applied on both weighted and unweighted datasets to ascertain risk levels and categorize them into six classes. Additionally, the classification task was executed using alternative algorithms. The outcomes of cross-validation and comparative analyses revealed that the performance was commendable for up to three classes, after which a decrease was observed in certain Naive Bayes and SVM classifiers. Among the Naïve Bayes family, the Bernoulli NB algorithm exhibited superior performance but was surpassed by Decision Trees, SVM, XGB, and Random Forests. Notably, the weighted dataset consistently outperformed the unweighted counterpart, with the allocation of weights significantly influencing algorithmic efficacy. The 80:20 split strategy yielded the most favorable outcomes in contrast to the 70:30 and 60:40 splits, albeit no significant variances were detected during cross-validation. Non-Naïve Bayes algorithms demonstrated superior performance compared to Naïve Bayes algorithms. For Naïve Bayes, optimal performance is achieved with three classes, highlighting its utility in conditional risk calculation, while non-Naïve Bayes multi-classification algorithms are more suitable for classification tasks due to the problem’s inherent compatibility with conditional probabilities. In conclusion, it is imperative to acknowledge that the characteristics of the data, the use of weights, and the data splitting methodology significantly influence the accuracy of machine learning algorithms in multi-class user classification.

Predicting early return to the operating room in early-onset scoliosis patients using machine learning techniques.

Surgical treatment of early-onset scoliosis (EOS) is associated with high rates of complications, often requiring unplanned return to the operating room (UPROR). The aim of this study was to create and validate a machine learning model to predict which EOS patients will go on to require an UPROR during their treatment course. A retrospective review was performed of all surgical EOS patients with at least 2 years follow-up. Patients were stratified based on whether they had experienced an UPROR. Ten machine learning algorithms were trained using tenfold cross-validation on an independent training set of patients. Model performance was evaluated on a separate testing set via their area under the receiver operating characteristic curve (AUC). Relative feature importance was calculated for the top-performing model. 257 patients were included in the study. 146 patients experienced at least one UPROR (57%). Five factors were identified as significant and included in model training: age at initial surgery, EOS etiology, initial construct type, and weight and height at initial surgery. The Gaussian naïve Bayes model demonstrated the best performance on the testing set (AUC: 0.79). Significant protective factors against experiencing an UPROR were weight at initial surgery, idiopathic etiology, initial definitive fusion construct, and height at initial surgery. The Gaussian naïve Bayes machine learning algorithm demonstrated the best performance for predicting UPROR in EOS patients. Heavier, taller, idiopathic patients with initial definitive fusion constructs experienced UPROR less frequently. This model can be used to better quantify risk, optimize patient factors, and choose surgical constructs. Prognostic: III.

Prediction of electric power performance of the exhaust waste heat recovery system of an automobile with thermoelectrical generator under real driving conditions by means of machine learning algorithms

In internal combustion engines, approximately 40% of the thermal power obtained from fuel burning is thrown out to the environment from the exhaust system. Implementations of waste heat recovery systems with thermoelectrical generator over exhaust system have become widespread as it is the waste heat resource with the highest temperature in a vehicle. During literature research, experiments related to waste heat recovery under real road conditions are very few and no study on estimation of system performance by machine learning algorithms has been found; therefore, Toyota Corolla has designed an air-cooled waste heat recovery system using 3 thermoelectrical generators for the exhaust system of the automobile. During the road tests, temperature and electric power generation values obtained as per gear, vehicle speed and engine speed have been recorded. In the study, a dataset consisting of 10 attributes was created with each record as a result of the path test. Using this dataset, Random Forest, Support Vector Machine (SVM) and Naive Bayes machine learning algorithms estimated the electrical power to be generated from the thermoelectric generator recovery system. In the study, 7 electric power classification estimates were made. In the estimation process, 76% of the dataset was used for training and 24% was used for testing. In terms of estimation success; an estimation success of 96.6% has been achieved by Random Forest method; 94.6% by Support Vector Machine (SVM) method; and 76.7% by Naive Bayes method. The results show that prospective electricity generation estimation can be achieved with high level of accuracy.

Predicting and determining antecedent factors of tourist village development using naive bayes and tree algorithm

This study aims to predict the progress status of tourism villages in the Kedung Ombo area, Java, Indonesia, and find the antecedent factors of the progress of tourism villages in Indonesia. This study uses a modern approach, namely data mining. Data sources for tourist villages use the data available on the Google link and the observation method. The prediction technique uses the Naïve Bayes machine learning algorithm and Tree Decision on Orange 3.3.0 software. The number of tourist villages analyzed was 126. The results showed that all tourist villages in the Kedung Ombo area were at the development level of the four tourist village classifications of the Ministry of Tourism and Creative Economy. The antecedent factors for the progress of tourism villages are the completeness of ICT facilities, multi-stakeholder partnerships, strong government support, community involvement, and various attractions. Another finding is that the Tree Decision algorithm provides better predictions than the Naïve Bayes method. The results of this study can be used to design policies for developing tourist villages throughout Indonesia.

Aerobic Fitness as an Important Moderator Risk Factor for Loneliness in Physically Trained Older People: An Explanatory Case Study Using Machine Learning.

Loneliness in older people seems to have emerged as an increasingly prevalent social problem. To apply a machine learning (ML) algorithm to the task of understanding the influence of sociodemographic variables, physical fitness, physical activity levels (PAL), and sedentary behavior (SB) on the loneliness feelings of physically trained older people. The UCLA loneliness scale was used to evaluate loneliness, the Functional Fitness Test Battery was used to evaluate the correlation of sociodemographic variables, physical fitness, PAL, and SB in the loneliness feelings scores of 23 trained older people (19 women and 4 men). For this purpose, a naive Bayes ML algorithm was applied. After analysis, we inferred that aerobic fitness (AF), hand grip strength (HG), and upper limb strength (ULS) comprised the most relevant variables panel to cause high participant loneliness with 100% accuracy and F-1 score. The naive Bayes algorithm with leave-one-out cross-validation (LOOCV) predicted loneliness in trained older with a high precision. In addition, AF was the most potent variable in reducing loneliness risk.

Violence Detection with Machine Learning: A Sociodemographic Approach

This study suggests that by implementing machine learning methods on a sociodemographic data set can be helpful in preventing domestic violence. This approach is important in predicting high-risk factors that an offender may cause and it offers treatment, and financial or mental health aids in order to prevent domestic violence. In this sense, this proposal is critical at a personal and social level in creating a secure and healthy environment as well as empowering an equal society. In our study, we use k-nearest neighbor (k-nn), support vector machine (SVM), decision tree (DT), and Gaussian Naive Bayes (GNB) machine learning algorithms for the prediction analysis. We provide the comparison of the classifiers with precision, recall, F1 score, and accuracy performance measures. According to our analysis, the decision tree (DT) performs the best performance in terms of accuracy.

COVID-19 HASTALIK TEŞHİSİNİN ADABOOST VE NAİVE BAYES ALGORİTMALARIYLA TAHMİN EDİLMESİ

Coronavirus disease (COVID-19) has infected millions of people all over the world and caused the death of many people. Identifying people with this disease as soon as possible is an important factor to prevent the disease from spreading. For disease detection, PCR (Polymerase Chain Reaction) tests performed. The results of tests always cannot give 100% accurate. In addition, obtaining information about test results sometimes may take a few days. Regarding the persons who applied to health institutions with suspicion of that illness, the diagnosis of COVID-19 disease takes place with the emergence of different disease symptoms. In this study, diagnostic estimates made for patients in the COVID-19 Surveillance dataset implementing Adaboost and Naive Bayes machine learning (ML) algorithm. It is possible to make predictions about new data by gaining experience from pre-existing data by means of using ML algorithms. In dataset determined within international disease codes for COVID-19 disease diagnosis estimates. Symptoms of patients used as attribute data in the dataset and used in binary format to be suitable for machine learning algorithms. According to the results obtained in this study, the classification forecast made with 85% accuracy with the Naive Bayes algorithm and 100% with the Adaboost algorithm.

Machine learning for the prediction of preclinical airway management in injured patients: a registry-based trial.

The aim of this study was to determine the feasibility of using machine learning to establish the need for preclinical airway management for injured patients based on a standardized emergency dataset. A registry-based, retrospective analysis was conducted of adult trauma patients who were treated by physician-staffed emergency medical services in southwestern Germany between 2018 and 2020. The primary outcome was to assess the feasibility of using the random forest (RF) and Naive Bayes (NB) machine learning algorithms to predict the need for preclinical airway management. The secondary outcome was to use a principal component analysis to determine the attributes that can be used and advanced for future model development. In total, 25,556 adults with multiple injuries were identified, including 1,451 patients (5.7%) who required airway management. Key attributes were auscultation, injury pattern, oxygen therapy, thoracic drainage, noninvasive ventilation, catecholamines, pelvic sling, colloid infusion, initial vital signs, preemergency status, and shock index. The area under the receiver operating characteristics curve was between 0.96 (RF; 95% confidence interval [CI], 0.96-0.97) and 0.93 (NB; 95% CI, 0.92-0.93; P<0.01). For the prediction of airway management, RF yielded a higher precision-recall area than NB (0.83 [95% CI, 0.8-0.85] vs. 0.66 [95% CI, 0.61-0.72], respectively; P<0.01). To predict the need for preclinical airway management in injured patients, attributes that are commonly recorded in standardized datasets can be used with machine learning. In future models, the RF algorithm could be used because it has robust prediction accuracy.

A Multiprotocol Controller Deployment in SDN-Based IoMT Architecture

Internet of Medical Things (IoMT) as a next-generation network requires heterogeneous services, technologies, and equipment infrastructure management resulting in more complex systems. The software-defined networking (SDN) approach has emerged as a promising solution to reduce this complexity by proposing a vendor-independent structure that disaggregates the control and data planes. In this study, an architecture based on the SDN is proposed for such heterogeneous and complex IoMT networks. A new controller that supports different wireless communication protocols has been developed for the control plane. We propose machine learning (ML)-based load balancing and time-sensitive prioritization (MLA) algorithms for dense and dynamic networks. An SDN-based IoMT network that consists of IEEE 802.15.6, TDMA, and IEEE 802.11 protocols is analyzed in a simulation program simultaneously using various scenarios in terms of throughput, delay, packet loss ratio, bit error rate, and user density parameters. In addition, in this study, a new data set is created for load balancing. The performances of support vector machine (SVM), ensemble of decision trees, k-NN, and Naive Bayes ML algorithms are compared, and SVM gives the best result with 95.1% accuracy.

BİLİŞİM SİSTEMLERİ ARAŞTIRMA ALANINDAKİ GENİŞLETİLMİŞ GERÇEKLİK YAYINLARININ METİN MADENCİLİĞİ VE DOĞAL DİL İŞLEME (NLP) TEKNİKLERİ YOLUYLA ANALİZİ

The aim of the study is to cluster and to classify the scientific papers regarding Extended Reality indexed in Web of Science database. To achieve this goal, Extended Reality related publications were located and gathered from the database. NLTK library was used for tokenization, stop words removal, and lemmatization operations. The TF-IDF vectorizer method in the Sklearn library was used to convert words to vector format. Then, the keywords of the publications were clustered using K-Means. The keywords in each cluster were searched throughout the abstract of each publication. The publication was labeled as the name of the cluster wherein the largest number of keywords matches the words in its abstract. Then, Support Vector Classifier, and Multinomial Naïve Bayes machine learning algorithms and Gated Recurrent Unit deep learning algorithms were conducted for classification. The results of deep learning and machine learning have been compared and this comparison yielded that the dataset is more suitable for deep learning in comparison to machine learning. Accuracy values are reported as 90.4%, 77.2%, and 99.8% for Support Vector Classifier, Multinomial Naïve Bayes, and Gated Recurrent Unit respectively. This study provides evidence that the GRU architecture is more effective than the classical machine learning algorithms.

Lung cancer prediction model using machine learning techniques

Lung cancer is cancer that forms in tissues of the lung, usually in the cells that line the air passages. It is the leading cause of cancer death in both men and women. Some of the Symptoms are Chest pain or discomfort, Trouble breathing, Wheezing, Blood in sputum (mucus coughed up from the lungs),Hoarseness, Loss of appetite, etc. Sometimes lung cancer does not cause any signs or symptoms. It may be found during a chest x-ray done for another condition. So early prediction of disease is very important to avoid death. So many machine learning algorithms are used to predict the lung cancer early but lack of accuracy. To overcome disease prediction accuracy issues, Gaussian Naive Bayes machine learning algorithm is used. The performance of the proposed GNB algorithm is evaluated using UCI Machine Learning Repository. The performance analysis shows GNB prediction model achieves 97.5%.

Naive bayes machine learning framework for auto detection of spam mails

Nowadays, Email spam is a major problem, with the increase usage of internet for product promotions, e-banking, e-loans, health related articles, movie news etc to users. Besides the people are practicing unwanted and unethical conduct as in phishing and mailing fraudulent information. Sending suspicious links through spam emails can harm our system and can also be sought after our system. Creating a fake user identity and email account is such easy for spammers. They pretend like a genuine entity in their spam emails and target those people who are not aware of these frauds. So, it is essential to identify those spam emails which are fraud. In this proposed work, we have applied machine learning algorithm to identify whether the incoming mail is spam or ham. This project will discuss how the Naive Bayes machine learning algorithm is applied on our data sets and how it has been selected for the email spam detection with the best precision and accuracy

The Mechanism of Spam Comment Detection Using Count Vectorizer and Naive Bayes Machine Learning Algorithms in Python

AbstractAim : The aim of this study is to predict Novel spam comments using Naive Bayes and SVM algorithms to improve the accuracy. Spamming is the process of posting unwanted and awkward comments on specific posts in any type of social sharing medium or video sharing medium. Materials and Methods : The Dataset used for training and testing of proposed text precision models is created using the comments of some popular videos from Youtube with 5 attributes and Sample size = 150. The framework that retrieves suitable data adapts SVM algorithm and compares it with Naive Bayes algorithm. Results and Discussion : The retrieval accuracy of SVM classifier is (96.72%) and Naive Bayes algorithm is (90.89%). There exists a statistical significant difference between SVM and Naive Bayes with (p&lt;0.05) Conclusion : The work has confirmed that the efficiency of the support vector machine algorithm has given more accuracy when compared to the Naive Bayes machine learning algorithm.

Stage related metabolic profile of the synovial fluid in patients with acute flares of knee osteoarthritis.

Osteoarthritis (OA) is the most common joint condition and the leading cause of pain and disability in elderly patients. Currently, there is no biomarker available for the early diagnosis of OA, and limited data is available regarding the molecular basis of progression for OA. For this reason, this study aimed to identify the metabolomic profile of early and late OA using high-performance liquid chromatography coupled with untargeted mass spectrometry (LC-MS). 31 patients with knee OA and joint effusion were enrolled. Based on Kellgren/Laurence scale, 12 patients were classified as early OA (eOA) and 19 as late OA (lOA). The synovial fluid (SF) was collected and characterized by untargeted LC-MS. Only the metabolites identified in more than 25% of each group were kept for further analysis. Principal component analysis (PCA) enabled the unsupervised clustering of the eOA and lOA groups. Further, for classification, the best three principal components (PCs) were used as input for two machine learning algorithms (random forest and naïve Bayes), which were trained to discriminate between the eOA and lOA groups. 43 metabolites were identified in both eOA and lOA, but after selecting the metabolites present in at least 25% of the patients in each group, the metabolomics analysis yielded a panel of only nine metabolites: four metabolites related to phospholipids (phosphatidylcholine 20:0/18:2 and 18:0/20:2, sphingomyelin, and ceramide), three metabolites belonging to purine metabolites (inosine 5'-phosphate, adenosine thiamine diphosphate, and diadenosine 5',5'-diphosphate), one metabolite was a gonadal steroid hormone (estrone 3-sulfate), and one metabolite represented by heme, with all but ceramide (d18:1/20:0) being enriched in the lOA group. By using as features the best three PCs (PC2, PC8 and PC9), random forest and naïve Bayes machine learning algorithms yielded a classification accuracy of 0.81 and 0.78, respectively. Our LC-MS analysis of SF from patients with eOA and lOA indicates stage-dependent differences, lOA being associated with a perturbed metabolome of phospholipids, purine metabolites, gonadal steroid hormones (estrone 3-sulfate) and a heme molecule. Specific questions need to be answered regarding the biosynthesis and function of these metabolites in osteoarthritic joints, with the aim of developing new relevant biomarkers and therapeutic strategies.

Classification of Non-Functional Requirements From IoT Oriented Healthcare Requirement Document.

Internet of Things (IoT) involves a set of devices that aids in achieving a smart environment. Healthcare systems, which are IoT-oriented, provide monitoring services of patients' data and help take immediate steps in an emergency. Currently, machine learning-based techniques are adopted to ensure security and other non-functional requirements in smart health care systems. However, no attention is given to classifying the non-functional requirements from requirement documents. The manual process of classifying the non-functional requirements from documents is erroneous and laborious. Missing non-functional requirements in the Requirement Engineering (RE) phase results in IoT oriented healthcare system with compromised security and performance. In this research, an experiment is performed where non-functional requirements are classified from the IoT-oriented healthcare system's requirement document. The machine learning algorithms considered for classification are Logistic Regression (LR), Support Vector Machine (SVM), Multinomial Naive Bayes (MNB), K-Nearest Neighbors (KNN), ensemble, Random Forest (RF), and hybrid KNN rule-based machine learning (ML) algorithms. The results show that our novel hybrid KNN rule-based machine learning algorithm outperforms others by showing an average classification accuracy of 75.9% in classifying non-functional requirements from IoT-oriented healthcare requirement documents. This research is not only novel in its concept of using a machine learning approach for classification of non-functional requirements from IoT-oriented healthcare system requirement documents, but it also proposes a novel hybrid KNN-rule based machine learning algorithm for classification with better accuracy. A new dataset is also created for classification purposes, comprising requirements related to IoT-oriented healthcare systems. However, since this dataset is small and consists of only 104 requirements, this might affect the generalizability of the results of this research.

A Remote Sensing Approach for Surface Urban Heat Island Modeling in a Tropical Colombian City Using Regression Analysis and Machine Learning Algorithms

The Surface Urban Heat Islands (SUHI) phenomenon has adverse environmental consequences on human activities, biophysical and ecological systems. In this study, Land Surface Temperature (LST) from Landsat and Sentinel-2 satellites is used to investigate the contribution of potential factors that generate the SUHI phenomenon. We employ Principal Component Analysis (PCA) and Multiple Linear Regression (MLR) techniques to model the main temporal and spatial SUHI patterns of Cartago, Colombia, for the period 2001–2020. We test and evaluate the performance of three different emissivity models to retrieve LST. The fractional vegetation cover model using Sentinel-2 data provides the best results with R2 = 0.78, while the ASTER Global Emissivity Dataset v3 and the land surface emissivity model provide R2 = 0.27 and R2 = 0.26, respectively. Our SUHI model reveals that the factors with the highest impact are the Normalized Difference Water Index (NDWI) and the Normalized Difference Build-up Index (NDBI). Furthermore, we incorporate a weighted Naïve Bayes Machine Learning (NBML) algorithm to identify areas prone to extreme temperatures that can be used to define and apply normative actions to mitigate the negative consequences of SUHI. Our NBML approach demonstrates the suitability of the new SUHI model with uncertainty within 95%, against the 88% given by the Support Vector Machine (SVM) approach.

Comparing shopping experiences in department stores and street markets: a big data analysis of TripAdvisor reviews

PurposeBased on big data analytical and statistical techniques, this study aims to examine tourists’ shopping experiences at department stores and street markets in Phuket.Design/methodology/approachA Naïve Bayes machine learning algorithm was used to identify the most frequently used terms in TripAdvisor reviews of both department stores and street markets contributed by the same pool of 729 tourists.FindingsA total of 18 out of 62 terms used were common in reviews of both shopping settings. However, the study found significant differences in the mean use of the 18 common terms and the likelihood of those terms being used in overall positive reviews.Practical implicationsThe study’s findings indicate differences in tourist shopping experiences at department stores and street markets. Several concrete recommendations are made, including a greater focus on the linkage to the national characteristic of street markets, and particularly the quality of local fruit, to enhance the tourist shopping experience.Originality/valueUnderstanding the differences between shopping malls and street markets from the tourist’s perspective would further enhance the coexistence of shopping malls and street markets in tourism-led growth cities. As such, using reviews of both shopping malls and street markets from an identical pool of tourists, the present study will analyse and compare tourists’ actual shopping experiences, thereby addressing this gap in the research canon via integrated statistical and big data analysis techniques.

Prediction on Domestic Violence in Bangladesh during the COVID-19 Outbreak Using Machine Learning Methods

The COVID-19 outbreak resulted in preventative measures and restrictions for Bangladesh during the summer of 2020—these unstable and stressful times led to multiple social problems (e.g., domestic violence and divorce). Globally, researchers, policymakers, governments, and civil societies have been concerned about the increase in domestic violence against women and children during the ongoing COVID-19 pandemic. In Bangladesh, domestic violence against women and children has increased during the COVID-19 pandemic. In this article, we investigated family violence among 511 families during the COVID-19 outbreak. Participants were given questionnaires to answer, for a period of over ten days; we predicted family violence using a machine learning-based model. To predict domestic violence from our data set, we applied random forest, logistic regression, and Naive Bayes machine learning algorithms to our model. We employed an oversampling strategy named the Synthetic Minority Oversampling Technique (SMOTE) and the chi-squared statistical test to, respectively, solve the imbalance problem and discover the feature importance of our data set. The performances of the machine learning algorithms were evaluated based on accuracy, precision, recall, and F-score criteria. Finally, the receiver operating characteristic (ROC) and confusion matrices were developed and analyzed for three algorithms. On average, our model, with the random forest, logistic regression, and Naive Bayes algorithms, predicted family violence with 77%, 69%, and 62% accuracy for our data set. The findings of this study indicate that domestic violence has increased and is highly related to two features: family income level during the COVID-19 pandemic and education level of the family members.

Invisible experience to real-time assessment in elite tennis athlete training: Sport-specific movement classification based on wearable MEMS sensor data

This study examined the reliability of a tennis stroke classification and assessment platform consisting of a single low-cost MEMS sensor in a wrist-worn wearable device, smartphone, and computer. The data that was collected was transmitted via Bluetooth and analyzed by machine learning algorithms. Twelve right-handed male elite tennis athletes participated in the study, and each athlete performed 150 strokes. The results from three machine learning algorithms regarding their recognition and classification of the real-time data stream were compared. Stroke recognition and classification went through pre-processing, segmentation, feature extraction, and classification with Support Vector Machine (SVM), including SVM without normalization, SVM with Min–Max, SVM with Z-score normalization, K-nearest neighbor (K-NN), and Naive Bayes (NB) machine learning algorithms. During the data training process, 10-fold cross-validation was used to avoid overfitting and suitable parameters were found within the SVM classifiers. The best classifier was achieved when C = 1 using the RBF kernel function. Different machine learning algorithms’ classification of unique stroke types yielded highly reliable clusters within each stroke type with the highest test accuracy of 99% achieved by SVM with Min–Max normalization and 98.4% achieved using SVM with a Z-score normalization classifier.

The development of a machine learning algorithm to identify occupational injuries in agriculture using pre-hospital care reports

PurposeCurrent injury surveillance efforts in agriculture are considerably hampered by the limited quantity of occupation or industry data in current health records. This has impeded efforts to develop more accurate injury burden estimates and has negatively impacted the prioritization of workplace health and safety in state and federal public health efforts. This paper describes the development of a Naïve Bayes machine learning algorithm to identify occupational injuries in agriculture using existing administrative data, specifically in pre-hospital care reports (PCR).MethodsA Naïve Bayes machine learning algorithm was trained on PCR datasets from 2008–2010 from Maine and New Hampshire and tested on newer data from those states between 2011 and 2016. Further analyses were devoted to establishing the generalizability of the model across various states and various years. Dual visual inspection was used to verify the records subset by the algorithm.ResultsThe Naïve Bayes machine learning algorithm reduced the volume of cases that required visual inspection by 69.5 percent over a keyword search strategy alone. Coders identified 341 true agricultural injury records (Case class = 1) (Maine 2011–2016, New Hampshire 2011–2015). In addition, there were 581 (Case class = 2 or 3) that were suspected to be agricultural acute/traumatic events, but lacked the necessary detail to make a certain distinction.ConclusionsThe application of the trained algorithm on newer data reduced the volume of records requiring visual inspection by two thirds over the previous keyword search strategy, making it a sustainable and cost-effective way to understand injury trends in agriculture.