Prediction Datasets Research Articles

Prediction of Human Physiological States by Using an Enhanced Recursive Feature Elimination Method

Human Action Recognition (HAR) is a vital part of the healthcare sector. Medical practitioners are experiencing difficulty in recognizing human physiological conditions due to the enormous quantity of sensory stimulation. Machine learning techniques help to predict human physiological conditions, providing medical practitioners to work more efficiently. Feature selection is an essential part of discovering new knowledge in the majority of real-world problems when there are a lot of features. Feature selection is extremely useful because it speeds up decisions and enhances classification performance. The importance of feature selection in machine learning is dimension reduction in a massive multivariate data collection. This paper presents an effective feature selection method known as the Enhanced Recursive Feature Elimination (EFRE) for selecting key features from the data set for HAR prediction. The experimental results reveal that the ERFE technique selects the most suitable features for HAR prediction. The performance of the proposed ERFE approach is tested using different performance evaluation metrics. The performance analysis shows that the ERFE method outperforms existing feature selection methods with 88% accuracy.

Molecular sharing and molecular-specific representations for multimodal molecular property prediction

Molecular property prediction plays a crucial role in drug discovery and development. However, traditional experimental measurements and Quantitative Structure-Activity Relationship (QSAR) models are often expensive, time-consuming, and data acquisition is challenging. To overcome these limitations and challenges, this study innovatively proposes a fusion molecular property prediction method called molecular property prediction model (MSSP) to address the non-uniqueness of Simplified Molecular Input Line Entry System (SMILES) string representation and the difficulty of capturing global information in molecular graphs. This method extracts multiple fingerprint features and utilizes graph neural network encoding to map different modalities of molecules into molecular sharing and molecular-specific representation spaces, achieving modal alignment and fusion of molecules by combining molecular invariance and representation specificity. To enhance the interpretability and visualization capabilities of the model, graph attention mechanisms are introduced, enabling the identification and inference of important chemical fragments within molecules. Experimental results on publicly available cell line phenotype and kinase activity datasets demonstrate that MSSP outperforms the current state-of-the-art methods in molecular property prediction. Additionally, MSSP exhibits strong competitiveness across nine benchmark molecular property prediction datasets. Furthermore, in the task of predicting SRC kinase data properties, this study successfully screens promising therapeutic compounds from compound libraries by validating the predictions of the MSSP model and combining them with traditional methods such as molecular docking and molecular dynamics simulations. Multiple potential Lyn inhibitors have been discovered through this approach. The application of MSSP model is helpful to discover new molecules with new drug properties or functions, accelerate the process of drug discovery, save time and resources, and provide important guidance for drug discovery.

Deep learning for geological mapping in the overburden area

This paper aims to achieve bedrock geologic mapping in the overburden area using big data, distributed computing, and deep learning techniques. First, the satellite Bouguer gravity anomaly with a resolution of 2′×2′ in the range of E66°-E96°, N40°-N55° and 1:5000000 Asia-European geological map are used to design a dataset for bedrock prediction. Then, starting from the gravity anomaly formula in the spherical coordinate system, we deduce the non-linear functional between rock density ρ and rock mineral composition m, content p, buried depth h, diagenesis time t and other variables. We analyze the feasibility of using deep neural network to approximate the above nonlinear generalization. The problem of solving deep neural network parameters is transformed into a non-convex optimization problem. We give an iterative, gradient descent-based solution algorithm for the non-convex optimization problem. Utilizing neural architecture search (NAS) and human-designed approach, we propose a geological-geophysical mapping network (GGMNet). The dataset for the network consists of both gravity anomaly and a priori geological information. The network has fast convergence speed and stable iteration during the training process. It also has better performance than a single neural network search or human-designed architectures, with the mean pixel accuracy (MAP) = 63.1% and the frequency weighted intersection over union (FWIoU) = 42.88. Finally, the GGMNet is used to predict the rock distribution of the Junggar Basin.

Integration of Adasyn Method with Decision Tree Algorithm in Handling Imbalance Class for Loan Status Prediction

Determining the provision of credit is generally carried out based on measuring credibility using credit analysis principles (5C principles). However, this method requires quite a long processing time and is very susceptible to subjective judgments which might influence the final results. This research uses data mining techniques by developing modeling on loan status prediction datasets. The stages in this research include data preprocessing, modeling, and evaluation using accuracy metrics and ROC graphs. In this analysis, it is known that there is a class imbalance in the processed dataset, so an oversampling technique must be carried out. This research uses the ADASYN (Adaptive Synthetic) Oversampling technique to ensure the class distribution is more balanced. Then, the ADASYN technique is integrated with the Decision Tree Algorithm to build a prediction model. The research results show that the two methods can increase prediction accuracy by 12.22%, from 73,91% to 85.22%. This improvement was obtained by comparing the accuracy results before and after using the ADASYN Oversampling technique. This finding is important because it proves that implementing such integration modeling can significantly improve the performance of classification models and provide strong potential for practical application in helping more effective loan status predictions.

Hybrid multimodal fusion for graph learning in disease prediction

Graph neural networks (GNNs) have gained significant attention in disease prediction where the latent embeddings of patients are modeled as nodes and the similarities among patients are represented through edges. The graph structure, which determines how information is aggregated and propagated, plays a crucial role in graph learning. Recent approaches typically create graphs based on patients' latent embeddings, which may not accurately reflect their real-world closeness. Our analysis reveals that raw data, such as demographic attributes and laboratory results, offers a wealth of information for assessing patient similarities and can serve as a compensatory measure for graphs constructed exclusively from latent embeddings. In this study, we first construct adaptive graphs from both latent representations and raw data respectively, and then merge these graphs via weighted summation. Given that the graphs may contain extraneous and noisy connections, we apply degree-sensitive edge pruning and kNN sparsification techniques to selectively sparsify and prune these edges. We conducted intensive experiments on two diagnostic prediction datasets, and the results demonstrate that our proposed method surpasses current state-of-the-art techniques.

Harnessing the potential of shared data in a secure, inclusive, and resilient manner via multi-key homomorphic encryption

In this manuscript, we develop a multi-party framework tailored for multiple data contributors seeking machine learning insights from combined data sources. Grounded in statistical learning principles, we introduce the Multi-Key Homomorphic Encryption Logistic Regression (MK-HELR) algorithm, designed to execute logistic regression on encrypted multi-party data. Given that models built on aggregated datasets often demonstrate superior generalization capabilities, our approach offers data contributors the collective strength of shared data while ensuring their original data remains private due to encryption. Apart from facilitating logistic regression on combined encrypted data from diverse sources, this algorithm creates a collaborative learning environment with dynamic membership. Notably, it can seamlessly incorporate new participants during the learning process, addressing the key limitation of prior methods that demanded a predetermined number of contributors to be set before the learning process begins. This flexibility is crucial in real-world scenarios, accommodating varying data contribution timelines and unanticipated fluctuations in participant numbers, due to additions and departures. Using the AI4I public predictive maintenance dataset, we demonstrate the MK-HELR algorithm, setting the stage for further research in secure, dynamic, and collaborative multi-party learning scenarios.

Low-cycle fatigue life prediction of austenitic stainless steel alloys: A data-driven approach with identification of key features

A data-driven approach for predicting the low-cycle fatigue (LCF) life of austenitic stainless steel (AusSS) is proposed to address the limitations of classical fatigue life prediction models. This approach integrates four key characteristics of the AusSS alloys, namely chemical compositions, microstructure, material processing parameters, and experimental parameters, to compile a comprehensive fatigue life prediction dataset. By employing the classification and regression technique, the relationship between various input features and LCF life is analyzed in depth, leading to the eventual identification of eight key features. Subsequently, various machine learning (decision tree, random forest, XGBoost, and support vector regression) and artificial neural network (ANN) are utilized to predict the LCF life using the compiled dataset. Comparative analysis of the results reveals higher prediction accuracy of ANN, with respective R2 and mean squared error values of 0.941 and 0.024, showcasing superior performance over the classical Basquin-Coffin-Manson model. This proposed methodology effectively captures the intricate feature-life relationships in AusSS, potentially reducing experimental time and costs of fatigue life estimation and offering promising applications in alloy design and manufacturing optimization.

Progressive Adapting and Pruning: Domain-Incremental Learning for Saliency Prediction

Saliency prediction (SAP) plays a crucial role in simulating the visual perception function of human beings. In practical situations, humans can quickly grasp saliency extraction in new image domains. However, current SAP methods mainly concentrate on training models in single domains, which do not effectively handle diverse content and styles present in real-world images. As a result, it would be of great significance if SAP models could efficiently adjust to new image domains. To this end, this article aims to design SAP models that can imitate the incremental learning ability of human beings on multiple image domains and name domain-incremental saliency prediction (DISAP). To make a tradeoff between preventing the forgetting of historical domains and achieving high performance on new domains, we propose a progressively updated domain incremental encoder. This encoder consists of a domain-sharing branch and a domain-specific branch. The domain-sharing branch includes a feature selection mechanism to preserve crucial parameters after fine-tuning the model on each current domain. The remaining parameters are reserved to absorb knowledge from future domains. Furthermore, to capture the unique characteristics of each domain with relatively low computational overhead, we introduce a lightweight design to construct the domain-specific branch, enabling effective adaptation to new domains. Extensive experiments are conducted on multiple domain-incremental learning settings formed by four saliency prediction datasets, including Salicon, MIT1003, the art subset of CAT2000, and WebSal. The results demonstrate that our method outperforms existing methods significantly. The code is available at https://github.com/KaIi-github/DIL4SAP .

Making More with Less: Improving Software Testing Outcomes Using a Cross-Project and Cross-Language ML Classifier Based on Cost-Sensitive Training

As digitalization expands across all sectors, the economic toll of software defects on the U.S. economy reaches up to $2.41 trillion annually. High-profile incidents like the Boeing 787-Max 8 crash have shown the devastating potential of these defects, highlighting the critical importance of software testing within quality assurance frameworks. However, due to its complexity and resource intensity, the exhaustive nature of comprehensive testing often surpasses budget constraints. This research utilizes a machine learning (ML) model to enhance software testing decisions by pinpointing areas most susceptible to defects and optimizing scarce resource allocation. Previous studies have shown promising results using cost-sensitive training to refine ML models, improving predictive accuracy by reducing false negatives through addressing class imbalances in defect prediction datasets. This approach facilitates more targeted and effective testing efforts. Nevertheless, these models’ in-company generalizability across different projects (cross-project) and programming languages (cross-language) remained untested. This study validates the approach’s applicability across diverse development environments by integrating various datasets from distinct projects into a unified dataset, using a more interpretable ML technique. The results demonstrate that ML can support software testing decisions, enabling teams to identify up to 7× more defective modules compared to benchmark with the same testing effort.

A digital pathology AI model to predict immune-oncology biomarker status and pembrolizumab response in a real-world cohort of patients with colorectal and breast cancer.

e13617 Background: Deep learning (DL) algorithms have garnered significant attention for prediction of biomarker status in digitized tissue slides. Here, we trained and validated a digital pathology algorithm to predict microsatellite instability (MSI)/mismatch repair deficiency (MMRd) in colorectal cancer (CRC) and PD-L1 status in breast cancer (BC) using a large real-world patient cohort. Methods: H&E and immunohistochemistry (IHC) slides of CRC and BC from Caris Life Sciences (Phoenix, AZ) with known biomarker status were digitally scanned (CRC: N=27,494, 6.6% MSI-high by next-generation sequencing [NGS]; N=25,788, 6.4% MMRd by IHC; BC: N=16,069, 15.8% PD-L1+ by IHC). A transformer-based DL model (Wagner, Sophia J., et al., 2023) was trained to predict biomarker status, with 5-fold cross-validation. Accuracy, F1 score, area under the curve (AUC), and other metrics were calculated comparing the model to NGS/IHC “truth”. Risk stratification was performed on holdout datasets of pembrolizumab-treated CRC (N=422-454) and triple negative breast cancer (TNBC; N=513) pts. Post-treatment overall survival (OS) was calculated from start of pembrolizumab to last contact, per insurance claims data. Hazard ratios (HR) and 95% confidence intervals (CI) were calculated by Cox-Proportional hazards and compared between model and “truth”. Results: When applied to holdout datasets for biomarker prediction, the validated model achieved AUC of 0.78-0.95. In CRC and BC, performance was better among primary specimens than metastatic. For MSI-H/MMRd risk stratification in pembrolizumab-treated CRC, HR for OS were comparable between model and “truth” for primary and metastatic cohorts. When applied to the TNBC holdout dataset, the model performed better than “truth” for PD-L1 risk stratification in pembrolizumab-treated pts. Conclusions: A digital pathology DL model predicts biomarker status with high accuracy and is non-inferior to NGS/IHC for risk stratification of pembrolizumab-treated CRC and TNBC pts. These results support further investigation of AI algorithms developed from large datasets as possible rapid screening tools for biomarker detection in molecular testing programs. [Table: see text]

LCSEP: A Large-Scale Chinese Dataset for Social Emotion Prediction to Online Trending Topics

LCSEP: A Large-Scale Chinese Dataset for Social Emotion Prediction to Online Trending Topics

Comparative analysis of explainable AI models for predicting lung cancer using diverse datasets

Lung cancer prediction is crucial for early detec-tion and treatment, and explainable AI models have gained attention for their interpretability. This study aims to compare various explainable AI models using diverse datasets for lung cancer prediction. Clinical, genomic, and imaging data from multiple sources were collected, prepro-cessed, and used to train models such as Logistic Regression, SVC-Linear, SVC-rbf, Decision Tree, Random Forest, AdaBoost Classifier, and XGBoost Classifier. Preliminary results indicate that Random Forest achieved the highest accuracy of 98.9% across multiple datasets. Evaluation metrics such as accuracy, precision, recall, and F1 score were utilized, along with interpretability techniques like feature importance rankings and rule extraction methods. The study's findings will aid in identifying effective and interpretable AI models, facilitating early detection and treatment decisions for lung cancer.

A novel hybrid deep learning model for early stage diabetes risk prediction

Diabetes is a prevalent global disease that significantly diminishes the quality of life and can even lead to fatalities due to its complications. Early detection and treatment of diabetes are crucial for mitigating and averting associated risks. This study aims to facilitate the prompt and straightforward diagnosis of individuals at risk of diabetes. To achieve this objective, a dataset for early stage diabetes risk prediction from the University of California Irvine (UCI) database, widely utilized in the literature, was employed. A hybrid deep learning model comprising genetic algorithm, stacked autoencoder, and Softmax classifier was developed for classification on this dataset. The performance of this model, wherein both the model architecture and all hyperparameters were specifically optimized for the given problem, was compared with commonly used methods in the literature. These methods include K-nearest neighbor, decision tree, support vector machine, and convolutional neural network, utilizing tenfold cross-validation. The results obtained with the proposed method surpassed those obtained with other methods, with higher accuracy rates than previous studies utilizing the same dataset. Furthermore, based on the study’s findings, a web-based application was developed for early diabetes diagnosis.

DKCNN: Improving deep kernel convolutional neural network-based COVID-19 identification from CT images of the chest.

An efficient deep convolutional neural network (DeepCNN) is proposed in this article for the classification of Covid-19 disease. A novel structure known as the Pointwise-Temporal-pointwise convolution unit is developed incorporated with the varying kernel-based depth wise temporal convolution before and after the pointwise convolution operations. The outcome is optimized by the Slap Swarm algorithm (SSA). The proposed Deep CNN is composed of depth wise temporal convolution and end-to-end automatic detection of disease. First, the datasets SARS-COV-2 Ct-Scan Dataset and CT scan COVID Prediction dataset are preprocessed using the min-max approach and the features are extracted for further processing. The experimental analysis is conducted between the proposed and some state-of-art works and stated that the proposed work effectively classifies the disease than the other approaches. The proposed structural unit is used to design the deep CNN with the increasing kernel sizes. The classification process is improved by the inclusion of depth wise temporal convolutions along with the kernel variation. The computational complexity is reduced by the introduction of stride convolutions are used in the residual linkage among the adjacent structural units.

A machine learning approach using stone volume to predict stone-free status at ureteroscopy.

To develop a predictive model incorporating stone volume along with other clinical and radiological factors to predict stone-free (SF) status at ureteroscopy (URS). Retrospective analysis of patients undergoing URS for kidney stone disease at our institution from 2012 to 2021. SF status was defined as stone fragments < 2mm at the end of the procedure confirmed endoscopically and no evidence of stone fragments > 2mm at XR KUB or US KUB at 3months follow up. We specifically included all non-SF patients to optimise our algorithm for identifying instances with residual stone burden. SF patients were also randomly sampled over the same time period to ensure a more balanced dataset for ML prediction. Stone volumes were measured using preprocedural CT and combined with 19 other clinical and radiological factors. A bagged trees machine learning model with cross-validation was used for this analysis. 330 patients were included (SF: n = 276, not SF: n = 54, mean age 59.5 ± 16.1years). A fivefold cross validated RUSboosted trees model has an accuracy of 74.5% and AUC of 0.82. The model sensitivity and specificity were 75% and 72.2% respectively. Variable importance analysis identified total stone volume (17.7% of total importance), operation time (14.3%), age (12.9%) and stone composition (10.9%) as important factors in predicting non-SF patients. Single and cumulative stone size which are commonly used in current practice to guide management, only represented 9.4% and 4.7% of total importance, respectively. Machine learning can be used to predict patients that will be SF at the time of URS. Total stone volume appears to be more important than stone size in predicting SF status. Our findings could be used to optimise patient counselling and highlight an increasing role of stone volume to guide endourological practice and future guidelines.

Pedestrian Trajectory Prediction Based on an Intention Randomness Influence Strategy

Pedestrian trajectory prediction is a key technical prerequisite for autonomous vehicle trajectory planning. However, a pedestrian is a changeable individual, and their intentions exhibit certain degrees of randomness and uncertainty, which leads to the issue that modeling only past trajectories does not enable the effective description of the random intentions and future trajectory directions of the pedestrian. Therefore, this paper proposes a flexible and embeddable stochastic intention vector construction strategy for modeling sudden pedestrian intention changes in real scenes and for better fitting the stochastic properties of pedestrian behaviors. First, we dynamically fuse historical trajectory information with random factors and construct an intention change probability based on the historical trajectory fitting errors of pedestrians, aiming to explicitly model the associated direction and velocity changes caused by random pedestrian intentions. Second, a new intention loss function is designed to guide the model to adaptively learn the probability of intention changes, which is used to dynamically describe pedestrian intention changes. Our proposed method is generalizable and can be applied as an embeddable module to any baseline pedestrian trajectory prediction method. The experimental results obtained on multiple large-scale public pedestrian trajectory prediction datasets demonstrate that our strategy achieves consistent performance improvements over different baselines.

Analysing an imbalanced stroke prediction dataset using machine learning techniques

Analysing an imbalanced stroke prediction dataset using machine learning techniques

Successive model-agnostic meta-learning for few-shot fault time series prognosis

Meta learning is a promising technique for solving few-shot fault prediction problems, which have attracted the attention of many researchers in recent years. Existing meta-learning methods for time series prediction, which predominantly rely on random and similarity matching-based task partitioning, facing three major limitations: (1) feature exploitation inefficiency; (2) suboptimal task data allocation; and (3) limited robustness with small samples. To this end, we introduce a novel‘meta-task’ partitioning scheme, underpinned by a differential autoregressive algorithm, that treats a continuous time period of a time series as a meta-task, composed of multiple successive short time periods. This approach allows us to extract more comprehensive features and relationships from the data, resulting in more accurate predictions. Furthermore, our findings indicate that the differential autoregressive approach significantly bolsters robustness across diverse datasets. Extensive experiments on several fault and time series prediction datasets demonstrate that our approach substantially enhances prediction performance and generalization capability under both few-shot and general conditions.

Improved Dwarf Mongoose Optimization Algorithm for Feature Selection: Application in Software Fault Prediction Datasets

Improved Dwarf Mongoose Optimization Algorithm for Feature Selection: Application in Software Fault Prediction Datasets

Japanese tort-case dataset for rationale-supported legal judgment prediction

AbstractThis paper presents the first dataset for Japanese Legal Judgment Prediction (LJP), the Japanese Tort-case Dataset (JTD), which features two tasks: tort prediction and its rationale extraction. The rationale extraction task identifies the court’s accepting arguments from alleged arguments by plaintiffs and defendants, which is a novel task in the field. JTD is constructed based on annotated 3477 Japanese Civil Code judgments by 41 legal experts, resulting in 7978 instances with 59,697 of their alleged arguments from the involved parties. Our baseline experiments show the feasibility of the proposed two tasks, and our error analysis by legal experts identifies sources of errors and suggests future directions of the LJP research.