Medical image analysis is a field where the classification of brain tumors is essential to allow proper diagnosis and treatment plan. This study suggests an effective mechanism of classifying brain MRI images into glioma, meningioma, pituitary and non-tumor images by combining deep learning and machine learning techniques. This model uses the AlexNet framework to extract deep features, Principal Component Analysis (PCA) to reduce dimensionalities and ReliefF to select features. To identify the best model, the chosen features are tested with the help of different supervised classifiers, such as Support Vector machine (SVM), K-Nearest Neighbor (KNN), Random Forest, and Naive Bayes. The proposed hybrid framework is evaluated on publicly available dataset. Optimal performance is obtained with the SVM classifier, with an accuracy of 94.36 percent and F1-score of 0.94, as the proposed approach is effective in improving the feature discriminability whilst incurring less computation.

Falls are a major safety concern, especially for older adults, and there is a continued need for reliable and efficient automatic fall detection systems. This paper presents a fall detection approach based on machine learning classifiers trained on numerical features extracted from the UR Fall Detection Dataset (URFD). Instead of operating directly on image or video data, the system uses eight compact geometric and motion-related features, including height-width ratio, major-minor ratio, bounding box occupancy, maximum standard deviation, and related descriptors. Four classification models are evaluated: Random Forest, Support Vector Machine (SVM), XGBoost, and LightGBM. All features are standardized using Standard Scaler prior to training. On a dataset of 11,542 samples, the best-performing model, tuned LightGBM, achieves an accuracy of 96.9%. The untuned LightGBM model reaches 96.4% accuracy, followed by XGBoost (95.6%), Random Forest (95.3%), and SVM (85.6%). The models are assessed using accuracy, precision, recall, and F1-score on a held-out test set of 2,309 samples. The results show that a feature-level representation combined with gradient boosting methods can provide competitive performance while remaining computationally efficient and more interpretable than deep learning approaches based on raw images.

The classification of brain tumors is an unsolved problem associated with heterogeneity of tumors and fluctuations in imaging conditions. In this work, the investigation introduces a powerful novel framework, named LHW-Net that combines handcrafted features called local binary patterns (LBP), histogram of oriented gradients (HOG), and wavelet transform (WT). Within the LHW-Net framework, the extracted features are utilized in different machine learning classifiers, such as K-Nearest Neighbors (KNN), Random Forest (RF) and Support Vector Classifier (SVC). The results of the individual classifiers are further combined using probabilistic score fusion approach to improve classification performance. The effectiveness and robustness of the proposed work are validated by the achieved experimental results on commonly accepted benchmark datasets.

The advent of DNA microarray datasets has revolutionized both machine learning (ML) and healthcare by providing critical insights into variations in gene expression across different tissue and cell samples. These datasets, which typically contain thousands to tens of thousands of features, are instrumental in aiding disease diagnosis and classification. However, many genes within these datasets are irrelevant or noisy, making classification tasks challenging due to the high dimensionality of the data and the limited sample sizes. Feature selection is a crucial strategy to mitigate these challenges, as it helps identify the most relevant genes while discarding the irrelevant ones. This process not only improves the performance of ML models but also plays a pivotal role in accurately classifying tumor types and refining treatment strategies. This chapter reviews recent advancements in feature selection methods specifically tailored for microarray data analysis. It highlights the common challenges, including data imbalance, complexity, and missing values, which can hinder the classification process. Additionally, it provides an empirical analysis of various feature selection techniques applied to representative datasets, offering a comparative perspective rather than identifying a single “best” method. Importantly, the interpretation of gene signatures resulting from feature selection also has potential applications in computational social science, such as identifying key factors influencing public health, social behaviors, and policy decisions.

Early detection of plant diseases is vital to ensuring agricultural productivity and safeguarding global food security. If left unmanaged, plant diseases can severely reduce crop yield and quality, posing risks to both farmers’ livelihoods and the broader societal food supply. Automated disease detection through machine learning, particularly using convolutional neural networks (CNN), enables farmers to take timely remedial actions. This chapter focuses on the use of CNN-based models for detecting plant leaf diseases in their early stages, providing actionable insights into disease management. CNNs are well-suited for this task because they can accurately process and analyze image data, identifying disease symptoms that might be difficult to detect with the naked eye. By focusing on early-stage detection, CNN-based models enable farmers to take timely corrective actions, reducing the spread of diseases and preventing significant crop damage. By experimentally evaluating different deep learning approaches, this chapter highlights how advanced machine learning techniques can significantly improve the accuracy of plant disease diagnosis. The results underscore the potential of these models not only to enhance agricultural practices but also to mitigate the societal impact of crop loss, contributing to sustainable food production and security.

The field of computational social science classification provides potent instruments for extracting insights from extensive social datasets. This chapter delves into the algorithms, practical applications, validation methodologies, and real-world instances of classification within social science research. It begins by emphasizing the pivotal role of classification in uncovering patterns and interconnections within social phenomena. This chapter delves deeply into computational classification, exploring prevalent algorithms at its core. It introduces supervised learning methods like logistic regression and decision trees, as well as unsupervised techniques such as k-means clustering. The discussion also addresses the selection of the most appropriate algorithm for a given research question. Additionally, it explores a range of applications across social science disciplines, such as community detection in social networks, sentiment analysis of online opinions, and image classification to analyze social behavior. Furthermore, it emphasizes the pivotal aspect of model validation, presenting essential evaluation metrics to assess classification performance and discussing cross-validation techniques to ensure generalizability. This chapter also addresses the critical issue of bias and fairness in classification models. Furthermore, the chapter illustrates the potency of classification through real-world case studies, including detecting political polarization on social media, forecasting customer churn, and categorizing news articles by topic. These case studies vividly exhibit the tangible application of classification in addressing a spectrum of social science challenges. By delving into algorithms, applications, validation techniques, and concrete examples, this chapter empowers social scientists with the insights necessary to harness computational classification for impactful research endeavors.

With cyber threats becoming more complex, Cyber- Attack Detection Systems (CADS) are crucial for identifying and preventing attacks. However, traditional methods often fail to detect advanced threats accurately, leading to problems with accuracy, scalability, and reliability. This chapter presents an ensemble- based machine learning framework, referred to Adaptive Weighted Voting Ensemble (AWVE) Model, that leverages the strengths of multiple models to enhance the detection and classification of cyber- attacks. By combining diverse predictive techniques within a unified system, the ensemble approach achieves improved performance, overcoming the shortcomings of individual models. The framework is evaluated using the NSL-KDD dataset, a well- known dataset for network intrusion detection research, to assess its effectiveness in identifying and categorizing various attack types. The results demonstrate that the ensemble framework achieves superior detection rates and classification accuracy across different attack categories while maintaining low false positive rates.

Skin Melanoma is a lethal type of cancer. The early diagnosis of which is crucial to improve the survival rate of the patients. Convolution neural networks are at the heart of the deep learning algorithms. In the present work authors have experimentally compared 2D and 3D Convolution Neural Network (CNN) models to identify the melanoma. We have employed three different types of datasets namely PH2, ISIC archive, and ISIC skin cancer datasets. We applied the two models on each of the datasets to determine their accuracy, precision, recall, f1 score and ROC curves. The experimental results provide the insights about the advantages and limitations of using 2D and 3D CNN models for the identification of skin melanoma. The authors have observed that 2D CNN model shows enhanced capabilities to detect skin lesion structures compared to 3D CNN. Moreover, the classification accuracy of the 2D CNN is also found better than 3D CNN.

Crop pests pose a great threat to global food security; thus, the best pest prevention measures must be implemented. By using different machine learning (ML) techniques to perform crop pest classification, this research provides ways to improve the accuracy and speed of identifying pests in agricultural sectors. Conventional methods for identifying pests frequently depend on manual observation, which is tedious, error-prone, and labor-intensive. On the other hand, machine learning (ML) presents an effective way to automate this procedure by using sophisticated techniques to analyze massive data sets and produce precise predictions. The study applies a variety of machine learning approaches, such as Random Forests, K-Nearest Neighbor, and Naive Bayes, to classify agricultural pests according to features that have been extracted from images. For model training and validation, an extensive collection of high-resolution images of different agricultural pests taken in a range of environmental settings is used. Metrics like accuracy are used to determine how well the machine learning models perform. The potential of machine learning approaches to revolutionize pest management in agriculture is evident from the results, which indicate how accurately they can identify and classify agricultural pests. The suggested method improves the overall effectiveness of pest management procedures and drastically reduces the time and effort required to identify pests. Ultimately, this research promotes more resilient and productive farming systems by supporting efforts to develop sustainable and technologically advanced solutions for addressing agricultural difficulties. The results demonstrate the potential of machine learning (ML) as an invaluable tool for farmers, agronomists, and policymakers, encouraging a proactive and data-driven approach to pest management in contemporary agriculture.

According to the World Health Organisation, falling is a major health problem with potentially fatal implications. Each year, thousands of people die as a result of falls, with seniors making up 80% of these fatalities. The automatic detection of falls may reduce the severity of the consequences. Our study focuses on developing a vision-based fall detection system. Our work proposes a new feature descriptor that results in a new fall detection framework. The body geometry of the subject is analyzed and patterns that help to distinguish falls from non-fall activities are identified in our proposed method. An AlphaPose network is employed to identify 17 keypoints on the human skeleton. Thirteen keypoints are used in our study, and we compute two additional keypoints. These 15 keypoints are divided into five segments, each of which consists of a group of three non-collinear points. These five segments represent the left hand, right hand, left leg, right leg and craniocaudal section. A novel feature descriptor is generated by extracting the distances from the segmented parts, angles within the segmented parts and the angle of inclination for every segmented part. As a result, we may extract three features from each segment, giving us 15 features per frame that preserve spatial information. To capture temporal dynamics, the extracted spatial features are arranged in the temporal sequence. As a result, the feature descriptor in the proposed approach preserves the spatio-temporal dynamics. Thus, a feature descriptor of size (Formula presented.) is formed where m is the number of frames. To recognize fall patterns, machine learning approaches such as decision trees, random forests, and gradient boost are applied to the feature descriptor. Our system was evaluated on the UPfall dataset, which is a benchmark dataset. It has shown very good performance compared to the state-of-the-art approaches.

Skeleton-based action recognition has become popular with the recent developments in sensor technology and fast pose estimation algorithms. The existing research works have attempted to address the action recognition problem by considering either spatial or temporal dynamics of the actions. But, both the features (spatial and temporal) would contribute to solve the problem. In this paper, we address the action recognition problem using 3D skeleton data by introducing eight Joint Distance Maps, referred to as Spatio Temporal Joint Distance Maps (ST-JDMs), to capture spatio temporal variations from skeleton data for action recognition. Among these, four maps are defined in spatial domain and remaining four are in temporal domain. After construction of ST-JDMs from an action sequence, they are encoded into color images. This representation enables us to fine-tune the Convolutional Neural Network (CNN) for action classification. The empirical results on the two datasets, UTD MHAD and NTU RGB+D, show that ST-JDMs outperforms the other state-of-the-art skeleton-based approaches by achieving recognition accuracies 91.63% and 80.16%, respectively.

With the recent developments in sensor technology and pose estimation algorithms, skeleton based action recognition has become popular. Classical machine learning methods based on hand-crafted features fail on large scale datasets due to their limited representation power. Recently, recurrent neural networks (RNN) based methods focus on the temporal evolution of body joints and neglect the geometric relations between them. In this paper, we propose eleven quadrilaterals to capture the geometric relations among joints for action recognition. An end-to-end 3-layer Bi-LSTM network is designed as Base-Net to learn robust representations. We propose two subnets based on the Base-Net to extract discriminative spatio temporal features. Specifically, the first subnet (SQuadNet) uses four spatial features and the second one (TQuadNet) uses two temporal features. The empirical results on two benchmark datasets, NTU RGB+D and UTD MHAD, show how our method achieves state of the art performance when compared to recent methods in the literature.

Human action recognition is a hot research topic in the field of computer vision. The availability of low cost depth sensors in the market made the extraction of reliable skeleton maps of human objects easier. This paper proposes three subnets, referred to as SNet, TNet, and BodyNet to capture diverse spatio-temporal dynamics for action recognition task. Specifically, SNet is used to capture pose dynamics from the distance maps in the spatial domain. The second subnet (TNet) captures the temporal dynamics along the sequence. The third net (BodyNet) extracts distinct features from the fine-grained body parts in the temporal domain. With the motivation of ensemble learning, a hybrid network, referred to as HNet, is modeled using two subnets (TNet and BodyNet) to capture robust temporal dynamics. Finally, SNet and HNet are fused as one ensemble network for action classification task. Our method achieves competitive results on three widely used datasets: UTD MHAD, UT Kinect and NTU RGB+D.

With the recent developments in sensor technology and pose estimation algorithms, skeleton based action recognition has become popular. This paper proposes a deep learning framework for action recognition task using ensemble learning. We design two subnets to capture spatial and temporal dynamics of the entire video sequence, referred to as Spatial -distance Net (SdNet) and Temporal - distance Net (TdNet) respectively. More specifically, SdNet is a Convolutional Neural Network based subnet to capture spatial dynamics of joints within a frame and TdNet is a long short term memory based subnet to exploit temporal dynamics of joints between frames along the sequence. Finally, two subnets are fused as one ensemble network, referred to as Spatio-Temporal distance Net (STdNet) to explore both spatial and temporal information. The efficacy of the proposed method is evaluated on two widely used datasets, UTD MHAD and NTU RGB+D, and the proposed STdNet achieved 91.16% and 82.55% accuracies respectively.

Action recognition is a leading research topic in the field of computer vision. This paper proposes an effective method for action recognition task based on the skeleton data. Four features are proposed based on the joint differences from 3D skeleton data. From the differences of 3D coordinates of corresponding joints in successive frames, three maps are extracted related to x, y and z coordinates respectively and then these maps are encoded into 2D color images, named as Joint Difference Maps (JDMs). The fourth JDM is formed by mapping the individual x, y and z difference maps into red, green and blue values. Hence, the 3D action recognition problem is converted into 2D image classification problem. It enables us to fine tune CNNs to learn informative features for 3D action recognition problem. The proposed method achieved 79.30% recognition rate on UTD MHAD dataset.

This paper addresses the action recognition problem using skeleton data. In this work, a novel method is proposed, which employs five Distance Maps (DM), named as Spatio- Temporal Distance Maps (ST-DMs), to capture the spatio-temporal information from skeleton data for 3D action recognition. Among five DMs, four DMs capture the pose dynamics within a frame in the spatial domain and one DM captures the variations between consecutive frames along the action sequence in the temporal domain. All DMs are encoded into texture images, and Convolutional Neural Network is employed to learn informative features from these texture images for action classification task. Also, a statistical based normalization method is introduced in this proposed method to deal with variable heights of subjects. The efficacy of the proposed method is evaluated on two datasets: UTD MHAD and NTU RGB+D, by achieving recognition accuracies 91.63% and 80.36% respectively.

This paper presents an approach for 3D action recognition using vector quantization with pairwise joint distance maps. We name this approach as VQ-PJDM. The main problem for 3D action recognition using skeleton data is that dealing with the variable length of action sequences. We solve this problem by approximation of each action sequence as a codebook, which is the output of Vector Quantization (VQ) method. The codebook size is fixed for any length of the action sequence. After all actions in the data set are approximated by VQ method, the Pairwise Distance Joint Distance Maps(PJDM) are calculated from approximated actions. The voting classifier is employed for action classification. The empirical results on the UT Kinect dataset prove that the proposed method gives better results than that of state of the art.

Action recognition in video sequence is a very important and challenging problem yet. This paper presents an efficient feature extraction method for human action recognition for depth video sequence. For the video sequence acquired by depth sensor, all 3-D projections (xy, yz and zx) are calculated for each depth frame. For each projection view, the difference between each alternative frames have been considered to form the Depth Motion Map (DMM). Principle Component Analysis technique is applied to decrease the facet of DMM-feature. Sequential minimal optimization (SMO) is pre-owned to train the Support Vector Machine (SVM). The proposed approach is evaluated on MSR Action-3D data set and compared with the existing approaches. The empirical results convey that proposed approach achieves good results than the existing approaches.