Instances Of Class Research Articles

Few-Shot Object Detection for SAR Images via Context-Aware and Robust Gaussian Flow Representation

In recent decades, few-shot object detection in SAR imagery has gained prominence as a major research focus. The unique imaging mechanism of SAR causes the model to suffer from foreground–background imbalance and inaccurate extraction of class prototypes for novel class instances. Therefore, we propose an innovative few-shot object detection algorithm for SAR images via context-aware and robust Gaussian flow representation. First, we design the Context-Aware Enhancement module to address the foreground–context imbalance problem by refining representative support features into fine-grained prototypes, which are deeply fused with query features based on the prototype matching paradigm. Second, we devise the Manifold Class Distribution Estimation module to address the difficulty of class distribution estimation and the fluctuation of class centers of the sparse novel class. Furthermore, we formulate the Category-Balanced Difference Aggregation module to model the relationship between the base class and the novel class, addressing the sensitivity of the model to the variance of the novel class instances. Finally, we design the Cosine Decoupling Module so that the aggregated features are executed only on the classification branch without affecting the precise localization of the target. Experiments based on SAR-AIRcraft-1.0 and the self-constructed MSAR-AIR dataset indicate that the fine-grained detection and identification performance of the novel class of airplanes can reach 32.90% and 55.26%, respectively, in the 10-shot and 50-shot cases. In addition, our method enables cross-domain detection for different scenarios and sample types and exhibits excellent generalization performance in data-sparse scenarios.

SAF-IS: A spatial annotation free framework for instance segmentation of surgical tools.

Instance segmentation of surgical instruments is a long-standing research problem, crucial for the development of many applications for computer-assisted surgery. This problem is commonly tackled via fully-supervised training of deep learning models, requiring expensive pixel-level annotations to train. In this work, we develop a framework for instance segmentation not relying on spatial annotations for training. Instead, our solution only requires binary tool masks, obtainable using recent unsupervised approaches, and tool presence labels, freely obtainable in robot-assisted surgery. Based on the binary mask information, our solution learns to extract individual tool instances from single frames, and to encode each instance into a compact vector representation, capturing its semantic features. Such representations guide the automatic selection of a tiny number of instances (8 only in our experiments), displayed to a human operator for tool-type labelling. The gathered information is finally used to match each training instance with a tool presence label, providing an effective supervision signal to train a tool instance classifier. We validate our framework on the EndoVis 2017 and 2018 segmentation datasets. We provide results using binary masks obtained either by manual annotation or as predictions of an unsupervised binary segmentation model. The latter solution yields an instance segmentation approach completely free from spatial annotations, outperforming several state-of-the-art fully-supervised segmentation approaches.

AICpred: Machine Learning-Based Prediction of Potential Anti-Inflammatory Compounds Targeting TLR4-MyD88 Binding Mechanism

Toll-like receptor 4 (TLR4) has been implicated in the production of uncontrolled inflammation within the body, known as the cytokine storm. Studies that employ machine learning (ML) in the prediction of potential inhibitors of TLR4 are limited. This study introduces AICpred, a robust, free, user-friendly, and easily accessible machine learning-based web application for predicting inhibitors against TLR4 by targeting the TLR4-myeloid differentiation primary response 88 (MyD88) interaction. MyD88 is a crucial adaptor protein in the TLR4-induced hyper-inflammation pathway. Predictive models were trained using random forest, adaptive boosting (AdaBoost), eXtreme gradient boosting (XGBoost), k-nearest neighbours (KNN), and decision tree models. To handle imbalance within the training data, resampling techniques such as random under-sampling, synthetic minority oversampling technique, and the random selection of 5000 instances of the majority class were employed. A 10-fold cross-validation strategy was used to evaluate model performance based on metrics including accuracy, balanced accuracy, and recall. The XGBoost model demonstrated superior performance with accuracy, balanced accuracy, and recall scores of 0.994, 0.958, and 0.917, respectively, on the test. The AdaBoost and decision tree models also excelled with accuracies ranging from 0.981 to 0.992, balanced accuracies between 0.921 and 0.944, and recall scores between 0.845 and 0.891 on both training and test datasets. The XGBoost model was deployed as AICpred and was used to screen compounds that have been reported to have positive effects on mitigating the hyperinflammation-associated cytokine storm, which is a key factor in COVID-19. The models predicted Baricitinib, Ibrutinib, Nezulcitinib, MCC950, and Acalabrutinib as anti-TLR4 compounds with prediction probability above 0.90. Additionally, compounds known to inhibit TLR4, including TAK-242 (Resatorvid) and benzisothiazole derivative (M62812), were predicted as bioactive agents within the applicability domain with probabilities above 0.80. Computationally inferred compounds using AICpred can be explored as potential starting skeletons for therapeutic agents against hyperinflammation. These predictions must be consolidated with experimental screening to enhance further optimisation of the compounds. AICpred is the first of its kind targeting the inhibition of TLR4-MyD88 binding and is freely available at http://197.255.126.13:8080.

Weakly Supervised Instance Segmentation in Aerial Images via Comprehensive Spatial Adaptation

Weakly supervised instance segmentation (WSIS) only employs image-level supervision to identify instance class labels and create segmentation masks, drawing increasing attention. Currently, existing WSIS methods primarily focus on activating the most discriminative regions and then inferring the entire instance by analyzing inter-pixel relationships within those regions. However, these identification regions are typically concentrated in limited but critical regions or are mistakenly activated in the background region, making it challenging to address scale variations among instances. Furthermore, different aerial instances often appear in close proximity, resulting in the merging of multiple instances of the same class. To tackle these challenges, we propose a comprehensive approach called Comprehensive Spatial Adaptation Segmentation (CSASeg). Specifically, the self-adaptive spatial-aware enhancement network (SSE) identifies extensive regions by analyzing spatial consistency within the class semantic map. Then, we develop a multi-level projection field (MPF) module to significantly enhance instance-level discrimination through deep-to-shallow residual estimation. Additionally, a foreground enhancement module is incorporated into SSE to reduce background noise while enhancing foreground details, significantly increasing the effectiveness of instance analysis. Extensive experiments conduct on three challenging datasets, iSAID, NWPU VHR-10.v2, and SSDD, demonstrate the competitiveness of our proposed approach.

Characterizing global work-conserving scheduling tardiness with uniform instances on multiprocessors

AbstractSoft real-time multiprocessor systems need scheduling policies introducing small overheads and for which it is possible to give guarantees on tardiness (i.e., the maximum delay that might arise with respect to job deadlines) in order to assess their feasibility in specific applications. For these reasons, lightweight policies such as Global Earliest Deadline First, and First-in First-out are preferred. Much effort has been spent in literature to provide efficiently computable tardiness bounds for periodic task systems scheduled on multiprocessors, but still, no exact bounds are known and results are given for specific classes of instances. In this paper, we use a work-conserving policy to schedule uniform instances, namely synchronous and periodic task systems in which tasks have the same period length and the same job length. We analytically derive a tight bound on the maximum tardiness and we give the exact length of the schedule hyper-period, showing that the latter can be computed in time linear in the number of processors. This result provides a lower bound to tardiness for the more general class of instances and is intended to close the gap with the upper bound from below.

Class-Incremental Learning: A Survey.

Deep models, e.g., CNNs and Vision Transformers, have achieved impressive achievements in many vision tasks in the closed world. However, novel classes emerge from time to time in our ever-changing world, requiring a learning system to acquire new knowledge continually. Class-Incremental Learning (CIL) enables the learner to incorporate the knowledge of new classes incrementally and build a universal classifier among all seen classes. Correspondingly, when directly training the model with new class instances, a fatal problem occurs - the model tends to catastrophically forget the characteristics of former ones, and its performance drastically degrades. There have been numerous efforts to tackle catastrophic forgetting in the machine learning community. In this paper, we survey comprehensively recent advances in class-incremental learning and summarize these methods from several aspects. We also provide a rigorous and unified evaluation of 17 methods in benchmark image classification tasks to find out the characteristics of different algorithms empirically. Furthermore, we notice that the current comparison protocol ignores the influence of memory budget in model storage, which may result in unfair comparison and biased results. Hence, we advocate fair comparison by aligning the memory budget in evaluation, as well as several memory-agnostic performance measures.

Customer Churn Prediction Model Based on Adaptive Clustering Mixed-Sampling

Predicting the probability of customer churn is an important reference for formulating and implementing customer retention strategies. Compared with single classification method, ensemble learning method can obtain better generalization ability and ensemble learning-based customer churn prediction method has gradually become a research focus. However, the prediction data of customer churn is usually imbalanced in class, and there are problems of overlapping and class-imbalance, so the prediction effect of ensemble learning model is ineffective. Therefore, an ensemble learning method based on adaptive clustering mixed-sampling (CUS-Ensemble) is proposed. This method is based on Bagging and regarded the gradient boosting decision tree as classifier, Firstly, cleanse the date of the whole training set, Secondly, the clustering under-sampling method is used to sample majority class instances, so that the sampled majority class instances are slightly more than the minority class instances, then, the sampled majority class instances and all minority class instances form a new training subset and provide a data cleansing again, next, using Borderline-SMOTE to balance the cleaned subset, and input it into the decision tree. Finally, the output of several gradient boosting decision trees is integrated as the final prediction result. Experimental results on six imbalanced customer datasets show that, compared with EasyEnsemble and other comparison methods, the AUC and Recall value of this method are increased by 4% and 13.6% on average, which is helpful to reduce the losses caused by customer churn.

Learning to Detect Novel Species with SAM in the Wild

AbstractThis paper tackles the limitation of a closed-world object detection model that was trained on one species. The expectation for this model is that it will not generalize well to recognize the instances of new species if they were present in the incoming data stream. We propose a novel object detection framework for this open-world setting that is suitable for applications that monitor wildlife, ocean life, livestock, plant phenotype and crops that typically feature one species in the image. Our method leverages labeled samples from one species in combination with a novelty detection method and Segment Anything Model, a vision foundation model, to (1) identify the presence of new species in unlabeled images, (2) localize their instances, and (3) retrain the initial model with the localized novel class instances. The resulting integrated system assimilates and learns from unlabeled samples of the new classes while not “forgetting” the original species the model was trained on. We demonstrate our findings on two different domains, (1) wildlife detection and (2) plant detection. Our method achieves an AP of 56.2 (for 4 novel species) to 61.6 (for 1 novel species) for wildlife domain, without relying on any ground truth data in the background.

Random Forest Algorithm for Toddler Nutritional Status Classification Website

Accurate data processing is essential for classifying toddler nutritional status on a website platform. The Random Forest algorithm is particularly effective in this context due to its ability to manage large datasets and mitigate overfitting. This study leverages Flask as the web framework to ensure responsiveness and adaptability, optimizing the data processing experience for users. Using secondary data comprising 120,999 records, the research aims to answer: "What factors affect the accuracy of the Random Forest model in classifying toddler nutritional status?" Model evaluation yielded excellent performance metrics, with accuracy, precision, recall, and F1-score values of 99.91%, 100%, 100%, and 100%, respectively. These results highlight the informative attributes in the dataset, such as age, gender, and height, that enhance classification accuracy. The Flask-based website enables users, such as healthcare professionals and policymakers, to input essential data points and receive instant classification results, thereby supporting prompt and informed responses to nutritional health issues. This study confirms that the Random Forest algorithm, combined with an intuitive web interface, effectively classifies toddler nutritional status with high accuracy.

Diversity-Representativeness Replay and Knowledge Alignment for Lifelong Vehicle Re-identification

Lifelong vehicle re-identification (LVReID) aims to match a target vehicle across multiple cameras, considering non-stationary and continuous data streams, which fits the needs of the practical application better than traditional vehicle re-identification. Nonetheless, this area has received relatively little attention. Recently, methods for lifelong person re-identification (LPReID) have been emerging, with replay-based methods achieving the best results by storing a small number of instances from previous tasks for retraining, thus effectively reducing catastrophic forgetting. However, these methods cannot be directly applied to LVReID because they fail to simultaneously consider the diversity and representativeness of replayed data, resulting in biases between the subset stored in the memory buffer and the original data. They randomly sample classes, which may not adequately represent the distribution of the original data. Additionally, these methods fail to consider the rich variation in instances of the same vehicle class due to factors such as vehicle orientation and lighting conditions. Therefore, preserving more informative classes and instances for replay helps maintain information from previous tasks and may mitigate the model’s forgetting of old knowledge. In view of this, we propose a novel Diversity-Representativeness Dual-Stage Sampling Replay (DDSR) strategy for LVReID that constructs an effective memory buffer through two stages, i.e. , Cluster-Centric Class Selection and Diverse Instance Mining. Specifically, we first perform class-level sampling based on density in the clustered class-centered feature space and then further mine the diverse, high-quality instances within the selected classes. In addition, we introduce Maximum Mean Discrepancy loss to align the feature distribution between replay data and the new arrivals and apply L2 regularisation in the parameter space to facilitate knowledge transfer, thus enhancing the model’s generalization ability to new tasks. Extensive experiments demonstrate effective improvements of our method compared to current state-of-the-art lifelong ReID methods on the VeRi-776, VehicleID, and VERI-Wild datasets.

Anticancer Potential of Piperidine Containing Small Molecule Targeted Therapeutics

AbstractIn the past two decades, targeted anti‐cancer therapeutics have achieved remarkable success due to their exceptional advantages of selectivity towards cancer cells and safety. Targeted small molecule anti‐cancer therapies persisted in many barriers; majorly poor response to drug therapy. Piperidine, a heterocyclic moiety, exceeds twenty instances of other pharmaceutical classes and natural compounds in the form of alkaloids effective in anti‐cancer treatment. The current review focuses on recent advancements, mainly from 2017–2023, of piperidine‐containing small molecule development as anti‐cancer agents. Total 10 piperidine containing anti‐cancer drugs have been approved by USFDA since 2017 to till date and around 15 small molecule anti‐cancer inhibitors containing piperidine scaffold which are in their early discovery phase have been reviewed which are classified according to their biological target. It highlights the structural contribution of piperidine ring towards the enhancement of activity or pharmacokinetic properties of diverse biological target‐specific anti‐cancer inhibitors of angiogenesis, EGFR, VEGFR, ALK, AKT1, topoisomerase, CDK2 etc. The role of the piperidine ring in enhancing potency, selectivity and bioavailability of novel molecules has been discussed. This review will be helpful to researchers, especially medicinal chemists, for the designing of piperidine‐containing potent drugs for specific biological targets for cancer.

LocustLens: leveraging environmental data fusion and machine learning for desert locust swarm prediction.

The desert locust is one of the most destructive locusts recorded in human history, and it has caused significant food shortages, monetary losses, and environmental calamities. Prediction of locust attacks is complicated as it depends on various environmental and geographical factors. This research aims to develop a machine-learning model for predicting desert locust attacks in 42 countries that considers three predictors: soil moisture, maximum temperature, and precipitation. We developed the Global Locust Attack Database for 42 countries (GLAD42) by integrating TerraClimate's environmental data with locust swarm attack data from the Food and Agriculture Organization (FAO). To improve the usability of spatial data, reverse geocoding which is the process of converting geographic coordinates (longitude and latitude) into human-readable location names (such as countries and regions) was employed. This step enhances the clarity and interpretability of the data by providing meaningful geographic context. This study's initial dataset focused on instances where locust attacks were recorded (positive class). To ensure a comprehensive analysis, we also incorporated negative class instances, representing periods (specific years and months) in the same countries and regions where locust attacks did not occur. This research utilizes the benefits of lazy learners by employing the K-nearest neighbor algorithm (K-NN), which provides high accuracy and the benefit of no time-consuming retraining even if real-time updated data is periodically added to the system. This research also focuses on building an eco-friendly machine learning model by evaluating carbon emissions from ML models. The results obtained from LocustLens are compared with other machine learning models, including baseline-K-NN, decision trees (DT), Logistic regression (LR), AdaBoost Classifier, BaggingClassifier, and support vector classifier (SVC). LocustLens outperformed all competitors with an accuracy of 98%, while baseline-K-NN achieved 96%, SVC gave 91%, DT gave 97%, AdaBoost has accuracy of 91%, BaggingClassifier gave 94% and LR gave 83%, respectively. Carbon emissions from RAM and CPU electricity consumption are measured in kg gCO2. They are a minimum for AdaBoost Classifier equal to 0.02 and 0.07 for DT and a maximum of 9.03 for SVC. The carbon footprint of LocustLens is 4.87 kg gCO2.

Imaging phenotype evaluation from digital breast tomosynthesis data: A preliminary study

Background:Digital breast tomosynthesis (DBT) has been widely adopted as a supplemental imaging modality for diagnostic evaluation of breast cancer and confirmation studies. In this study, a deep learning-based method for characterizing breast tissue patterns in DBT data is presented. Methods:A set of 5388 2D image patches was produced from 230 right mediolateral oblique, 259 left mediolateral oblique, 18 right craniocaudal, and 15 left craniocaudal single-breast DBT studies, using slice-wise annotations of abnormalities and normal tissue. We implemented a patch classifier to predict samples according to two differing scenarios and train it using the patch dataset. First, tissue samples were classified into the following classes: malignant, benign, and normal breast tissue. Second, tissue samples were classified into the following classes: malignant mass, benign mass, benign architectural distortion, malignant architectural distortion, and normal breast tissue. We employed transfer learning and initialized the model base layers with existing pre-trained weights obtained from Globally-Aware Multiple Instance Classifier. Results:High class-wise recall values of 0.8906, 0.8541 and 0.7345 and specificities 0.9558, 0.9575 and 0.8830 were obtained for normal, benign, and malignant classification, respectively. More intricate classification yielded class-wise recall values of 0.8708, 0.8299, 0.9444 and 0.5723 and specificities 0.9406, 0.9833, 0.8943 and 0.9652 for benign mass, normal, malignant architectural distortion, and malignant mass, respectively. However, benign architectural distortion was confused with benign mass and malignant architectural distortion. Conclusions:Combining the proposed phenotype classifier with the commonly used malignant-benign-normal classification enables a more detailed assessment of digital breast tomosynthesis images.

Inverting Cryptographic Hash Functions via Cube-and-Conquer

MD4 and MD5 are fundamental cryptographic hash functions proposed in the early 1990s. MD4 consists of 48 steps and produces a 128-bit hash given a message of arbitrary finite size. MD5 is a more secure 64-step extension of MD4. Both MD4 and MD5 are vulnerable to practical collision attacks, yet it is still not realistic to invert them, i.e., to find a message given a hash. In 2007, the 39-step version of MD4 was inverted by reducing to SAT and applying a CDCL solver along with the so-called Dobbertin’s constraints. As for MD5, in 2012 its 28-step version was inverted via a CDCL solver for one specified hash without adding any extra constraints. In this study, Cube-and-Conquer (a combination of CDCL and lookahead) is applied to invert step-reduced versions of MD4 and MD5. For this purpose, two algorithms are proposed. The first one generates inverse problems for MD4 by gradually modifying the Dobbertin’s constraints. The second algorithm tries the cubing phase of Cube-and-Conquer with different cutoff thresholds to find the one with the minimum runtime estimate of the conquer phase. This algorithm operates in two modes: (i) estimating the hardness of a given propositional Boolean formula; (ii) incomplete SAT solving of a given satisfiable propositional Boolean formula. While the first algorithm is focused on inverting step-reduced MD4, the second one is not area-specific and is therefore applicable to a variety of classes of hard SAT instances. In this study, 40-, 41-, 42-, and 43-step MD4 are inverted for the first time via the first algorithm and the estimating mode of the second algorithm. Also, 28-step MD5 is inverted for four hashes via the incomplete SAT solving mode of the second algorithm. For three hashes out of them, it is done for the first time.

Resampling approach for imbalanced data classification based on class instance density per feature value intervals

In practical applications, imbalanced datasets significantly degrade the classification performance of machine learning models. However, most conventional resampling approaches fall short in adequately addressing the varying contributions of individual features to the classification model. In response to this defect, this study introduces three novel resampling approaches. The first approach, Oversampling based on class instance density per feature value intervals (OCF), focuses on augmenting the dataset. The second approach, Undersampling based on class instance density per feature value intervals (UCF), seeks to reduce dataset size. The third approach, Hybrid sampling based on class instance density per feature value intervals (HSCF), which can perform oversampling and undersampling simultaneously. These approaches categorize feature value into different intervals based on their varying information content, calculate class instance densities within these intervals, and generate feature values in intervals with high discriminative information. Subsequently, these generated features are combined to synthesize minority class data, effectively achieving oversampling. Additionally, the study combines class instance density and feature importance to identify majority class data at the classification boundary with minimal contribution and subsequently executes undersampling. The flexibility to adjust sampling ratios and the integration of OCF and UCF enable the implementation of hybrid sampling. Finally, experiments on the benchmark dataset demonstrate the superiority and effectiveness of the proposed method. Furthermore, it is observed that the method proposed in this study enhances the feature dividing capability of decision tree classifiers. Hence, the best results are achieved when working in synergy with decision tree classifiers, leading to the most significant improvements in classification performance. All codes have been published at https://github.com/Wangfeiopen/HSCF.

Truthful two-facility location with candidate locations

We study a truthful two-facility location problem in which a set of agents have private positions on the line of real numbers and known approval preferences over two different facilities. Given the locations of the two facilities, the cost of an agent is the total distance from the facilities she approves. The goal is to decide where to place the facilities from a given finite set of candidate locations so as to (a) approximately optimize desired social objectives, and (b) incentivize the agents to truthfully report their private positions. We focus on the class of deterministic strategyproof mechanisms and show bounds on their approximation ratio in terms of the social cost (i.e., the total cost of the agents) and the max cost for several classes of instances depending on the preferences of the agents over the facilities.

BENN: Balanced Ensemble Neural Network for Handling Class Imbalance in Big Data

ABSTRACTClass imbalance is a critical challenge in big data analytics, often leading to biased predictive models. This imbalance can lead to biased models that perform well on the majority class but poorly on the minority class. Many machine learning models tend to be biased towards the majority class because they aim to minimise overall error, often leading to poor performance on the minority class. This paper presents the balanced ensemble neural network, a novel solution to effectively address class imbalance in big data. Balanced ensemble neural network combines the robust capabilities of neural networks with the power of ensemble learning, incorporating class balancing strategies to ensure fair representation of minority classes. The methodology involves integrating multiple neural networks, each trained on balanced subsets of data using techniques like Synthetic Minority Over‐sampling Technique and Random Undersampling. This integration aims to leverage the strengths of individual networks while reducing their inherent biases. Our extensive experiments across various datasets reveal that BENN achieves an AUC‐ROC score of 0.94, surpassing other models such as random forest (0.88), support vector (0.84) and single neural net (0.80). It was also observed that BENN's performance is better compared to traditional neural network models and standard ensemble methods in key metrics like accuracy, precision, recall, F1‐score and AUC‐ROC. The results specifically highlight BENN's effectiveness in accurately classifying instances of minority classes, a notable challenge in many existing models. These findings underscore BENN's potential as a substantial advancement in handling class imbalance within big data environments, offering a promising direction for future research and application in machine learning.

From shallows to depths: unveiling hybrid synthetic data modeling for enhanced learning with privacy considerations in naturally imbalanced datasets

Data serves as the foundational element that drives model development and performance in machine learning and deep learning. The learning algorithms are as good as data on which they are trained on. Data constrained environments pose serious threat to the effectiveness of learning algorithms. Data limitations stem from a range of factors, including regulatory restrictions, privacy concerns, and the inherent scarcity of relevant data. This constrained availability of data often leads to class imbalance problem in the context of classification tasks. To address the challenges of limited data, the algorithmic generated data, also known as synthetic data, is gaining significant traction as a cost-effective, readily available, and secure alternative. Synthetic data generation techniques can be employed to enhance dataset size by augmenting data samples and to address class imbalance by increasing the number of minority class instances. These techniques generally fall into two main categories: distance-based shallow models and probability estimation-based deep generative models. Shallow interpolation-based models generate new data points within the local space between existing data points, while deep density estimation based models generate new data by learning the whole distribution of data. . In the context of smaller datasets, deep generative models often struggle to accurately estimate the probability distribution of whole data. To effectively represent the global data distribution, these deep models require initial starting data samples to guide their approximation. This paper examines the potential of integration of shallow and deep generative models in the data generation pipeline for effective synthetic data augmentation which furthers enhanced learning and generalization of downstream tasks. In this work, we present a hybrid approach of tabular data generation involving mixed type data attributes (continuous, discrete) and pay special attention to data imbalance and insufficient data problems. We introduce the Hybrid Data Balancing and Augmentation Approach for Mixed Tabular Data (HDBA-MTD), specifically designed to synthesize samples for underrepresented labels of output class and address issues of insufficient data instances. This approach enhances training data diversity, thereby paying special attention to the downstream classification and generalization performance. Experiments are carried out using benchmark datasets to assess the practicality of the presented hybrid model in real-world scenarios. This work has also attempted to quantify the privacy preservability for real data concerning ethical considerations and data security circumstances. The evaluation and analysis of these experiments show that the present hybrid model performs favorably compared to other current hybrid synthetic data generation methods.

Rethinking Multiple Instance Learning for Whole Slide Image Classification: A Good Instance Classifier Is All You Need

Rethinking Multiple Instance Learning for Whole Slide Image Classification: A Good Instance Classifier Is All You Need

A hierarchical heterogeneous ant colony optimization based oversampling algorithm using feature similarity for classification of imbalanced data

Imbalanced data classification is one of the challenging problems in machine learning. Oversampling is a promising technique that generates synthetic minority instances to balance the dataset. Inappropriate minority instances generated may deteriorate the performance of the classifier. Majority of the oversampling algorithms create new minority instances by choosing nearest neighbors for random interpolation. However, these methods do not provide new information to the dataset and therefore standard classifiers do not show good performance on such datasets. Therefore, it is necessary to generate diverse minority class instances to increase the performance of the classifier. Since, every feature of each minority class instance contribute valuable information, generating synthetic instances from the features of all minority instances would produce diverse minority instances, thereby increasing the performance of the classifier. This paper proposes a Hierarchical Heterogeneous Ant Colony Optimization based oversampling algorithm using Feature Similarity (HHACO-FSOTe) for generation of synthetic minority instances. Instead of choosing few neighbors for interpolation, the proposal considers all minority instances for generation of synthetic instances. HHACO-FSOTe generates new feature values by computing the minimum absolute difference between the features of a given minority instance and the corresponding features of the remaining minority instances. The features in the dataset are distributed among the ant agents enabling parallelism, thereby reducing the time taken for oversampling. HHACO-FSOTe do not require parameter tuning or training. The proposal is evaluated on 41 low dimensional, 11 high dimensional and 8 noisy datasets. Experiments reveal that HHACO-FSOTe is competent with the state-of-art oversampling techniques. Results were validated using non-parametric statistical tests.