Search 250 million research papers

Classification Functions for Machine Learning and Data Mining

Tsutomu Sasao · Synthesis lectures on digital circuits and systems · 2023

Cited by 5Open access

Machine Learning and Terrain Classification from Ladar Data

2008

While there has been much research concerning ladar range data and terrain classification, the two topics rarely overlap. This paper presents a method for real-time terrain classification using a Naïve Bayes Decision Tree trained on features extracted from data gathered by a ladar sensor. The classification method involves breaking down a ladar image into voxels, classifying the terrain in each voxel, and then extrapolating an overall image classification from the component voxel classifications. The method for feature extraction from each voxel by Principal Component Analysis is described, as are the results of testing the classifier on real-world data. I.

Cited by 0

Crime Classification Using Machine Learning and Data Analytic

Muhammad Naqi Haider, Rafia Mumtaz, Syed Mohammad Hassan Zaidi · 2022

Crimes have both short-term and long-term effects on individuals and on society as a whole. Modern-day law enforcement agencies are making use of data analytics and machine learning algorithms for predictive policing in order to prevent crimes. There is a lack of crime reporting systems in developing countries like Pakistan. This needs improvement because the crime reporting systems serve as a backbone of generating crime data for analysis. To build a crime prediction system, it is advisable to analyze the minimalistic requirements in advance and use similar data sets from other countries for analysis and prediction. This can serve to provide the government and other authorities with insights into the benefits of correcting crime reporting systems and deploying crime prediction systems in the future. This study implements three machine learning algorithms namely Naïve Bayesian Classifier, Decision Tree Classifier, and Random Forest Classifier for predicting the primary crime type, to highlight potential data problems such as a high number of distinct values for features, class imbalance, and lack of independent features, which can be expected in Pakistan crime data in order to build crime prediction systems. The performance of the algorithms to predict crime type is analyzed on the datasets, where, the Random Forest Classifier performed outperformed Naïve Bayes and Decision tree with an accuracy of 55.03% respectively. In its entirety, this study demonstrates that the limitation of the feature set is the primary cause of the low accuracy of predictions and suggests the inclusion of more features to be extracted for improving the accuracy of predictions. Finally, this research work highlights the lack of structured crime data in the country and provides suggestions for the application of this study in Pakistan.

Cited by 4

Machine learning classification of geochemical and geophysical data

Leijun Huang, Mehala Balamurali, Katherine L. Silversides · 2019

Accurate classifications in downhole exploration data are essential for exploration, geological modelling and prediction of mining outputs. These holes are typically interpreted manually, which is slow, subjective and prone to error. Machine learning techniques can potentially automate these classifications. Our test deposit contains the shale dominated West Angelas (WA) Member, and banded iron formation and iron ore in the Mount Newman (MN) Member. Deep learning using autoencoders, Support Vector Machine (SVM) and k-means were applied to classify stratigraphy and rock type using mineral groups or geochemical assays. Autoencoder produced accuracies of 83.2–95.1%, K-means accuracies of 30.4–58.8%, and SVM accuracies of 81.6–88.5%. Geochemical assays were a better indicator of stratigraphy. Autoencoder and SVM produced significantly better results than k-means. This was probably due to k-means not using training data. Although they gave similar results, autoencoder is preferable to SVM for this application as it can handle more than two categories.

Cited by 2

Machine Learning for Biomedical Data Analysis and Classification

Abdullah Abd Al-Jaleel Majeed Al-Mustaqbal, Hassan Hussein Khudair, Ammar Kamel et al. · Bright Sky Publications eBooks · 2024

Buy Machine Learning for Biomedical Data Analysis and Classification written by Abdullah Abd Al-Jaleel Majeed Al-Mustaqbal, Hassan Hussein Khudair, Ali Salam Kamel, Hasaneien Abass Hussien and Mountazer Samir Obaid at low price.

Cited by 0Open access

Machine learning in football data prediction and classification

Pinto, Tiago Soeira · The Scientific Repository of the Polytechnic Institute of Porto (Polytechnic Institute of Porto) · 2024

Com a globalização do desporto e a crescente importância atribuída à análise de dados recolhidos durante o jogo, a avaliação do desempenho individual e coletivo no futebol tornouse uma prática comum. No entanto, esta análise frequentemente carece do uso de ferramentas baseadas em aprendizagem automática (ML). Nesse sentido, esta dissertação explora o uso de técnicas de ML e pretende alcançar dois objetivos principais. O primeiro consiste na resolução de um problema de regressão, destinado a prever a eficácia dos remates do jogador com base em indicadores de desempenho como a sua posição, os minutos jogados, o total de remates e os golos marcados. O segundo visa resolver um problema de classificação, que classifique os níveis de desempenho do jogador com base na eficácia dos seus remates. Para ambas as tarefas, foram examinados e comparados vários métodos de aprendizagem automática; a Árvore de Decisão e o Gradient Boosting foram considerados os mais eficazes. Estes modelos demonstraram resultados superiores na previsão da eficácia dos jogadores e na classificação do seu desempenho, oferecendo uma nova abordagem à análise do futebol que vai para além da análise estatística convencional.

Cited by 0Open access

Machine Learning Models and Algorithms for Big Data Classification

Shan Suthaharan · Integrated series on information systems/Integrated series in information systems · 2015

Cited by 648

Machine Learning for Classification and Clustering of Dementia Data

F Guest (21932024) · Open Research Exeter (University of Exeter) · 2021

Dementia is a term used to describe heterogeneous diseases that can generally be characterised by a decline in cognitive ability that affects daily living. It has been predicted that the prevalence of dementia will increase significantly over the coming years, thus it is a priority worldwide. This thesis discusses research conducted with two primary aims. They were to investigate the use of machine learning for distinguishing between people with and without dementia, as well as differentiating between key dementia subtypes where appropriate; and to gain an understanding of the inherent structure of dementia data, to ultimately investigate disease signatures. Data was acquired from the National Alzheimer's Coordinating Center in the United States, and a data set comprising 32,573 observations and 260 features of mixed type was utilised. It included features whose values were constrained by relations with others, as well as two types of missingness which arose when data was unexpectedly not recorded and when the information was irrelevant or unobtainable for a known reason, respectively. Notably, the former genuinely missing values were imputed where possible, whilst the latter conditionally missing values were handled. An imputation approach was developed, which simultaneously builds a random forest classifier while handling conditionally missing values. It maintained the known relations in the data set, so far as possible. A clustering approach was also developed that ultimately measures the similarity of observations based on the similarity of their paths through the trees of an isolation forest before employing spectral clustering. Crucially, it can naturally draw on variables of mixed type. A dementia classifier with an area under the receiver operating characteristic curve (AUC) of 0.99 and 10 pairwise dementia subtype classifiers with AUCs ranging from 0.88 to 1.0 (rounded) were produced, suggesting machine learning could be a useful tool for diagnosing dementia and differentiating between the main subtypes. Key features were identified using these classifiers and were markedly different for the two types of diagnosis. Furthermore, preliminary experiments conducted using the clustering approach suggested that mild cognitive impairment may be a mild form of dementia as opposed to a clinical entity, over which there is much debate; and there could be evidence for the current subtypes. Ultimately, these findings have the potential to transform the way dementia is diagnosed.

Cited by 0Open access

Terpene classification using mass spectra data and machine learning

Shah, Deepkumar · AcadiaU-DEV · 2022

Traditionally, separation and classification of terpenes from food samples is done with thehelp of two processes, namely: Gas Chromatography (GC) (separate terpenes from each other) and Mass Spectrometry (MS) (detects the separated terpenes). A mass spectrum is matched to known terpenes by a chemist using an existing database of various samples. We propose a machine learning solution that detects multiple terpenes from mass spectra without separating them. Synthetic data is generated using six common terpenes, cannabid-iol (CBD) and tetrahydrocannabinol (THC) using an additive composition model with five percent resolution in each component. The six base terpenes included in the current study are myrcene, linalool, limonene, α-pinene, β-pinene, and trans-caryophyllene, referred to as target chemicals. These terpenes are found in many food products that we consume in our day-to-day lives. There are 50,388 possible mass spectra combinations of these terpenes. The research investigates the development of a machine learning approach that classifies and quantifies the target chemicals. The system reads the mass spectra and automatically detects target chemicals that are present and determines their relative proportions. Various supervised and unsupervised classification and regression models are built and tested.Principal Component Analysis (PCA) of the eight target chemicals is performed to test how distinguishable the terpenes are. Classification of the major contributor target chemical with the help of artificial neural networks and a deep learning algorithm results in an accuracy of 99.1% for target chemicals respectively based on a train-test split methodology. Regression results with the help of ANNs and a single-layered model result in a mean absolute error of 0.001222 for target chemicals based on a train-test split methodology.

Cited by 0Open access

Intent Classification Using Machine Learning Algorithms and Augmented Data

Alaa T. Al-Tuama, Dhamyaa A. Nasrawi · 2022

An essential step in a conversational agent is an intent classification of user-generated text input. The purpose of building the intent classifier for a chatbot is to understand the intention of user queries to respond fast and accurately. Robust chatbots necessitate more utterances for an improved training model. Nevertheless, acquiring and annotating data is time-consuming and expensive. This paper investigates machine learning techniques and data augmentation for addressing intent classification. Experiments were conducted on office product's question answering of Amazon using Random Forest, Multinomial Naïve Bayes, Logistic Regression, and Support Vector Machine (SVM). Contextual word embedding with BERT was used for generating new synonym utterances. The main experiments are the comparison of the performance of these methods after augmenting new data. In general, SVM and random forest yield comparable results. Followed by logistic regression. However, the f1 score of the multinomial naïve base is the lowest. Additionally, we discovered that augmenting new utterances had a simple effect on the performance of models.

Cited by 13

Lung Cancer Classification Using Machine Learning and Data Augmentation

R. El Idrissi, Nabila Zrira, H. Ez‐Zahraouy et al. · Lecture notes in networks and systems · 2025

Cited by 0

Water Leak Classification Using Acoustic Data and Machine Learning

Abdelhafid Tobbal, Reda Kara, Billel Bengherbia et al. · 2024

Water Distribution Networks (WDNs) are critical infrastructures for delivering potable water from its source to consumers. Unfortunately, these systems often suffer significant water loss from leaks which typically remain undetected until they result in extensive damage. To effectively address these issues there is a pressing need for innovative maintenance strategies that enable early fault detection and rapid response. This paper presents a diagnostic approach that uses acoustic monitoring and Edge-AI techniques to detect failures in urban water pipelines through hydrophone sensor recordings. The methodology employed in this study involves extracting both time-domain and frequency-domain features to classify various pipeline faults using three distinct Machine Learning (ML) models: Artificial Neural Network (ANN), Random Forest (RF), and Support Vector Machine (SVM). Furthermore, we implemented the processing method on an embedded platform, demonstrating its effectiveness in real-world scenarios. The results of this study reveal an average fault detection accuracy of 96.83% with a fast response time of 1.055 seconds.

Cited by 1

Data complexity in machine learning and novel classification algorithms

Yaser S. Abu‐Mostafa, Ling Li · 2006

We investigate the role of data complexity in the context of binary classification problems. The universal data complexity is defined for a data set as the Kolmogorov complexity of the mapping enforced by that data set. It is closely related to several existing principles used in machine learning such as Occam's razor. We demonstrate its application in data decomposition and data pruning. In data decomposition, we illustrate that a data set is best approximated by its principal subsets which are Pareto optimal with respect to the complexity and the set size. In data pruning, we show that outliers usually have high complexity contributions, and propose methods for estimating the complexity contribution. We propose a family of novel learning algorithms to directly minimize the 0/1 loss for perceptrons. Our algorithms usually achieve the lowest training error and are also computationally efficient. Such advantages make them favorable for both standalone use and ensemble learning. Experiments show that our algorithms work very well with AdaBoost. AdaBoost is an ensemble methods that aggregate many base hypotheses for binary classification problems. It can be viewed as gradient descent on a margin cost function. We provide a more sophisticated abstract boosting algorithm, CGBoost, based on conjugate gradient in function space. When the AdaBoost exponential cost function is optimized, CGBoost generally yields much lower cost and training error but higher test error, which implies that the exponential cost is vulnerable to overfitting. With the optimization power of CGBoost, we can adopt more regularized cost functions that have better out-of-sample performance but are difficult to optimize. A multiclass classification problem can be reduced to a collection of binary problems with the aid of a coding matrix. A coding matrix with strong error-correcting ability may not be overall optimal if the binary problems are too hard for the base learner. In this paper, we propose a new multiclass boosting algorithm that modifies the coding matrix according to the learning ability of the base learner. We show experimentally that our algorithm is very efficient in optimizing the multiclass margin cost, and outperforms existing multiclass algorithms such as AdaBoost.ECC and one-vs-one.

Cited by 21

Machine Learning and Data Analysis Based Breast Cancer Classification

Souvik Das, Rama Chaitanya Karanam, Obilisetty Bala Krishna et al. · 2023

The precise classification of breast cancer significantly affects the further diagnosis of the cancer patient. Usual methods of testing and classifying of the cancer cells in the laboratory takes a long time. Laboratory work is so complicated and needs the expertise of highly qualified doctors. Instead, machine learning algorithms can be used on the data of cancer patients to classify breast cancer in relatively lesser time. The data are usually obtained from the images of the cancer cells. This way of classifying cancer provides more reliable results and saves a lot of time and money. This current study examines the cancer patient data obtained from images, preprocessing of the data, training of different machine learning models, and choosing of the best-trained model for the classification of breast cancer. The current study helps not only in identifying the best machine learning model to classify breast cancer but also to formulate a process of diagnosing breast cancer based on the preprocessed data obtained from the raw images of breast cancer.

Cited by 2

Music Genre Classification Using Data Mining and Machine Learning

Nimesh Ramesh Prabhu, James Andro-Vasko, Doina Bein et al. · Advances in intelligent systems and computing · 2018

Cited by 4

ECG data classification and explainability with machine learning and deep learning algorithms

Ojha, Jaya · Duo Research Archive (University of Oslo) · 2024

This thesis investigates the application of explainable artificial intelligence (XAI) algorithms for electrocardiogram (ECG) data classification, focusing on the performance of the enhanced ST-CNN-5 model compared to established models. Results indicate a slight improvement in accuracy, suggesting enhanced predictive capabilities. However, lower specificity and area-under-curve metrics underscore the need for further comprehensive evaluation. For explainability, Shapley Additive Explanations (SHAP), Gradient-weighted Class Activation Mapping (GradCAM), and Local Interpretable Model-agnostic Explanations (LIME) were utilized which reveals the superior ability of SHAP to highlight key ECG characteristics. The thesis also evaluates Support Vector Machine (SVM) and Random Forest (RF) algorithms, finding the enhanced ST-CNN-5 model outperforms in predictive accuracy. Minimal bias by sex is observed, emphasizing ongoing scrutiny of demographic variables. This research provides valuable insights into the effectiveness of algorithms and the interpretability of models, thereby aiding the development of AI-driven diagnostic tools in clinical settings.\n\nIn connection with the outcomes of this thesis, we have contributed a paper to the Intelligent Cross-Data Analysis and Retrieval conference, which was accepted for publication in April 2024. The appendix A section of the thesis contains a copy of the aforementioned paper for reference.\n\nAccepted paper in Conference:\n• Jaya Ojha, Hårek Haugerud, Anis Yazidi and Pedro G. Lind. Exploring Interpretable AI Methods for ECG Data Classification, ICDAR ’24: The 5th ACM Workshop on Intelligent Cross-Data Analysis and Retrieval\n\nFurthermore, we are currently in the process of preparing a scientific report based\non the findings of this research.

Cited by 0Open access

Product Review Classification using Machine Learning and Statistical Data Analysis

Kajal Singh · International Journal of Recent Technology and Engineering (IJRTE) · 2023

The aim of the paper is to implement and analyze the machine learning models for product review dataset. The project focuses on binary classification, multi-class classification, and clustering approaches to analyze and categorize product reviews. The performance of the models over each of the five classification tasks is measured by the 5-fold cross-validation scores over the training data.

Cited by 2Open access

Classification of Cattle Behavior Leveraging Accelerometer Data and Machine Learning

Ahmet Pütün, Derya Yılmaz · 2024

In recent years, with the increasing world population, the utilization of developing technologies in order to meet the sustainability needs in the livestock sector and to increase the efficiency in animal products and to ensure animal welfare has become an important field of study. In this study, in order to contribute to the improvement of livestock operational processes, animal behaviors were classified by using accelerometer sensor data obtained from cattle movements and machine learning methods. The accelerometer data collected from six cows (black Japanese cattle) in thirteen different behavior categories were analyzed. Different classification algorithms unlike the previous studies conducted with this data set were tried, and model performances were improved with the use of fewer features compared to previous ones. Statistical and entropy-based features were extracted from the accelerometer data and multi-class classification was performed to detect thirteen different behaviors using Extreme Gradient Boosting (XGBoost), Extra Trees (ET), Light Gradient Boosting Machine (LightGBM), Multilayer Perceptron (MLP) and Support Vector Machine (SVM) classifiers. When the classification performances were evaluated, it was seen that the ET Classifier model provided the best performance with an accuracy rate of 97.48%. The obtained result is above the highest accuracy rate obtained in the studies used this dataset in the literature and obtained using fewer features. This performance is also at a high level that can be compared with other studies in this fields. In conclusion, this study offers great potential for monitoring and management of animal behavior, and especially the determination of drinking and feeding behaviors with high accuracy will provide significant benefits in terms of monitoring and management of cattle health.

Cited by 1

Swift Imbalance Data Classification using SMOTE and Extreme Learning Machine

Rishabh Rustogi, Ayush Prasad · 2019

Continuous expansion in the fields of science and technology has led to the immense availability and attainability of data in every field. Fundamentally understanding and analyzing this data is a critical job in the decision-making process. Although, great success has been achieved by the prevailing data engineering and mining techniques, the problem of swift classification of the imbalanced data still exists in academia and industry. A potential solution to the problem of skewness in data can be resolved by data upsampling or downsampling. There exists a few techniques that firstly remove skewness and then perform classification, however, these methods suffer from hurdles like abortive precision or slower learning rate. In this paper, a hybrid method to classify binary imbalanced data using Synthetic Minority Over-sampling Technique followed by Extreme Learning Machine is proposed. Our method along with swift learning rate is efficacious to predict the desired class. We verified our model using five standard imbalance dataset and obtained higher F-measure, G-mean and ROC score for all the dataset.

Cited by 21

Machine Learning and Handcrafted Image Processing Weed Classification Results Data

Igathinathane Cannayen · Data Archiving and Networked Services (DANS) · 2023

Shape features extracted from ImageJ particle analysis based on four weed species (Common Lambsquarters, Common Purslane, Horseweed, and Redroot Pigweed) images that were used in the hand crafted image processing and machine learning models/methodology development to identify the weeds.

Cited by 0Open access