Range Of Machine Learning Approaches Research Articles

Flight Ticket Fare Prediction using Supervised Learning

Abstract: With the increasing interconnectivity of air routes worldwide, air travel has become a widespread and expeditious means of transportation. Forecasting airline fares poses a significant and intricate challenge due to their continual fluctation, influenced by a diverse array of factores. Extensive research suggests that the application of Machine Learning, and Deep Learing techniques enables the swift estimation of flight fares at specific intervals. This study employs a Machine Learning Regression methodology to predict flight fares based on essential parameters such as departure and arrival times, departure location, destination, stopovers, and airline provider. This research employs two distinct datasets for training and testing purposes, evaluating a range of machine learning approaches to forecast flight ticket prices.

Detection of Pulmonary Embolism: Workflow Architecture and Comparative Analysis of the CNN Models

Machine learning has proven to be a practical medical image processing technique for pattern discovery in low-quality labelled and unlabeled datasets. Deep vein thrombosis and pulmonary embolism are both examples of venous thromboembolism, which is a key factor in patient mortality and necessitates prompt diagnosis by experts. An immediate diagnosis and course of treatment are necessary for the life-threatening cardiovascular condition known as pulmonary embolism (PE). In the study of medical imaging, especially the identification of PE, machine learning (ML) algorithms have produced encouraging results. This study's objective is to assess how well machine learning (ML) algorithms perform in identifying PE in computed tomography (CT) scans. A range of ML approaches were used to the dataset, including deep learning algorithms such as convolutional neural networks. The effectiveness of PE detection systems can be greatly enhanced by the use of cutting-edge methodologies like deep learning, which lowers the possibility of incorrect diagnoses and enables the quick administration of therapy to individuals who require it. This work contributes to the growing body of evidence that supports the use of ML in medical imaging and diagnosis. Future research should examine how these algorithms might be included into clinical workflows, resolving any potential implementation challenges, and making sure their adoption is done so in a secure and efficient way. In this study, we provide a thorough evaluation of three different models: the streamlined architecture MobileNetV2 with an accuracy of 96%, compared to other models like the Xception model with an accuracy of 91%, and the Efficientnet B5 model with an accuracy of 97%, after observation and process following.

Cardiac ailment recognition using ML techniques in E-healthcare

Heart ailments can take numerous forms, and they are frequently referred to as cardio vascular illnesses. These can range from heart rhythm problems to birth anomalies to blood vessel disorders. It has been the main cause of death worldwide for several decades. To recognize the illness early and properly manage, it is critical to discover a precise and trustworthy approach for automating the process. Processing massive amounts of data in the field of medical sciences necessitates the application of data science. Here we employ a range of machine learning approaches to examine enormous data sets and aid in the accurate prediction of cardiac diseases. This paper explores the supervised learning models of Naive Bayes, Support Vector Machine, K-Nearest Neighbors, Decision Tree, in order to provide a comparison investigation for the most effective method. When compared to other algorithms, K-Nearest Neighbor provides the best accuracy at 86.89%.

Prediction of Antibody-Antigen Binding via Machine Learning: Development of Data Sets and Evaluation of Methods

Background The mammalian immune system is able to generate antibodies against a huge variety of antigens, including bacteria, viruses, and toxins. The ultradeep DNA sequencing of rearranged immunoglobulin genes has considerable potential in furthering our understanding of the immune response, but it is limited by the lack of a high-throughput, sequence-based method for predicting the antigen(s) that a given immunoglobulin recognizes. Objective As a step toward the prediction of antibody-antigen binding from sequence data alone, we aimed to compare a range of machine learning approaches that were applied to a collated data set of antibody-antigen pairs in order to predict antibody-antigen binding from sequence data. Methods Data for training and testing were extracted from the Protein Data Bank and the Coronavirus Antibody Database, and additional antibody-antigen pair data were generated by using a molecular docking protocol. Several machine learning methods, including the weighted nearest neighbor method, the nearest neighbor method with the BLOSUM62 matrix, and the random forest method, were applied to the problem. Results The final data set contained 1157 antibodies and 57 antigens that were combined in 5041 antibody-antigen pairs. The best performance for the prediction of interactions was obtained by using the nearest neighbor method with the BLOSUM62 matrix, which resulted in around 82% accuracy on the full data set. These results provide a useful frame of reference, as well as protocols and considerations, for machine learning and data set creation in the prediction of antibody-antigen binding. Conclusions Several machine learning approaches were compared to predict antibody-antigen interaction from protein sequences. Both the data set (in CSV format) and the machine learning program (coded in Python) are freely available for download on GitHub.

SPOSDS: A smart Polycystic Ovary Syndrome diagnostic system using machine learning

Polycystic Ovary Syndrome (PCOS) is a hormonal disorder that affects a large percentage of women of reproductive age. PCOS causes imbalanced or delayed menstrual cycles and produces high levels of the male hormone. The ovaries may create a significant number of little fluid-filled sacs (follicles) yet fail to discharge eggs regularly. The actual cause of PCOS is uncertain. However, early exposure and curing, as well as weight loss, may lower the threat of long-term complications. This study focuses on PCOS diagnosis based on a clinical dataset supplied by Kottarathil, accessible via its Kaggle repository. Non -invasive screening parameters are used to evaluate a range of machine learning approaches for screening PCOS patients without the use of invasive diagnostics. According to the findings of the experiments, the Random Forest (RF) method outperforms the other prominent machine learning algorithms with an accuracy of 93.25%. Further, the out-of-bag (OOB) error is utilized for assessing the prediction performance of RF.

Fast, Detailed, Accurate Simulation of a Thermal Car-Cabin Using Machine-Learning

Car-cabin thermal systems, including heated seats, air-conditioning, and radiant panels, use a large proportion of the energy budget of electric vehicles and thus reduce their effective range. Optimising these systems and their controllers might be possible with computationally efficient simulation. Unfortunately, state-of-the-art simulators are either too slow or provide little resolution of the cabin’s thermal environment. In this work, we propose a novel approach to developing a fast simulation by machine learning (ML) from measurements within the car cabin over a number of trials within a climatic wind tunnel. A range of ML approaches are tried and compared. The best-performing ML approach is compared to more traditional 1D simulation in terms of accuracy and speed. The resulting simulation, based on Multivariate Linear Regression, is fast (5 microseconds per simulation second), and yields good accuracy (NRMSE 1.8%), which exceeds the performance of the traditional 1D simulator. Furthermore, the simulation is able to differentially simulate the thermal environment of the footwell versus the head and the driver position versus the front passenger seat, but unlike a traditional 1D model cannot support changes to the physical structure. This fast method for obtaining computationally efficient simulators of car cabins will accelerate adoption of techniques such as Deep Reinforcement Learning for climate control.

Abstract LB129: Stratification of radiotherapy and fluoropyrimidine-based chemotherapy from multi-omic profiling in rectal cancer biopsies

Abstract Neoadjuvant chemoradiotherapy is commonly used to treat rectal cancer but patients have different levels of response and/or toxic effects. As part of the Stratification in COloRecTal cancer (S:CORT) programme, we collected 257 rectal biopsies from two cohorts: Grampian (single hospital) and Aristotle (clinical trial). All patients had been subsequently treated with identical regimen of neoadjuvant radiotherapy and capecitabine. We performed trancriptomic, mutation and copy number profiling and aimed to identify biomarkers associated with the robust pathological endpoint of complete response (CR). Key biological determinants were identified by linear regression of different pre-defined, hypothesis-driven biomarkers for radiotherapy response, adjusted by the known confounders T and N stage. A novel RNA signature was derived using a personalised bioinformatical pipeline using a wide range of machine learning approaches. Results were validated in a publicly available transcriptomic cohort of 107 patients treated with similar dose of radiotherapy and 5-fluorouracil infusion. Further comparision of the biological determinants and the novel RNA signature were performed in the same cohorts and also TCGA by linear regression. Previously published transcriptomic signatures were retrieved and assessed in the validation, unseen cohort. Grampian and Aristotle cohorts had similar statistical power and showed similar associations of CR with biological candidates, 10 of them being significant or borderline (p<0.1). Accordingly, both cohorts were merged into a single discovery set to better assess which ones would show additive, independent association. Following multivariable stepwise regression the final model was composed of the immune biomarkers cytotoxic lymphocytes and CMS1 for radiosensitivity while the stromal TGFb Fibroblasts and epithelial APC mutations were for radioresistance. The first three variables were validated in the transcriptomic validation set (Cyt lymph OR 7.09, p=0.01; CMS1 OR 5.39, p=0.02; TGFb Fib OR 0.27, p=0.04). In parallel, a 33-gene signature, trained in the discovery cohort by a comprehensive machine learning pipeline, showed excellent predictive ability in the validation cohort (0.9 AUC; 88% accuracy, 90% sensitivity, 86% specificity). Most genes were associated with at least one of the four biological features identified in the discovery set, validation set and a third cohort of colorectal cancer resections. Our novel signature showed much better predictive ability than other previously published transcriptomic signatures in the validation, unseen cohort. The immune, stromal and epithelial components of rectal tumours are important players for prediction of CR to radiotherapy in rectal cancer. A 33-gene transcriptomic biomarker can be used to effectively select patients that are highly likely to achieve CR allowing organ preservation while modulation of the relevant biological features in the other patients may be tested to improve their poor outcome with current treatment strategies. Citation Format: Enric Domingo, Sanjay Rathee, Andrew Blake, Leslie M. Samuel, Graeme I. Murray, David Sebag-Montefiore, Simon Gollins, Nicholas West, Rubina Begum, Marian Duggan, Laura White, Susan Richman, Philip Quirke, James Robineau, Keara Redmond, Aikaterini Chatzipli, Ultan McDermott, Ian Tomlinson, Philip Dunne, Francesca Buffa, Tim Maughan. Stratification of radiotherapy and fluoropyrimidine-based chemotherapy from multi-omic profiling in rectal cancer biopsies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr LB129.

The accuracy versus interpretability trade-off in fraud detection model

AbstractLike a hydra, fraudsters adapt and circumvent increasingly sophisticated barriers erected by public or private institutions. Among these institutions, banks must quickly take measures to avoid losses while guaranteeing the satisfaction of law-abiding customers. Facing an expanding flow of operations, effective banking relies on data analytics to support established risk control processes, but also on a better understanding of the underlying fraud mechanism. In addition, fraud being a criminal offence, the evidential aspect of the process must also be considered. These legal, operational, and strategic constraints lead to compromises on the means to be implemented for fraud management. This paper first focuses on the translation of practical questions raised in the banking industry at each step of the fraud management process into performance evaluation required to design a fraud detection model. Secondly, it considers a range of machine learning approaches that address these specificities: the imbalance between fraudulent and nonfraudulent operations, the lack of fully trusted labels, the concept-drift phenomenon, and the unavoidable trade-off between accuracy and interpretability of detection. This state-of-the-art review sheds some light on a technology race between black box machine learning models improved by post-hoc interpretation and intrinsic interpretable models boosted to gain accuracy. Finally, it discusses how concrete and promising hybrid approaches can provide pragmatic, short-term answers to banks and policy makers without swallowing up stakeholders with economical and ethical stakes in this technological race.

Dynamic Butterfly Optimization Algorithm for Feature Selection

Feature selection represents an essential pre-processing step for a wide range of Machine Learning approaches. Datasets typically contain irrelevant features that may negatively affect the classifier performance. A feature selector can reduce the number of these features and maximise the classifier accuracy. This paper proposes a Dynamic Butterfly Optimization Algorithm (DBOA) as an improved variant to Butterfly Optimization Algorithm (BOA) for feature selection problems. BOA represents one of the most recently proposed optimization algorithms. BOA has demonstrated its ability to solve different types of problems with competitive results compared to other optimization algorithms. However, the original BOA algorithm has problems when optimising high-dimensional problems. Such issues include stagnation into local optima and lacking solutions diversity during the optimization process. To alleviate these weaknesses of the original BOA, two significant improvements are introduced in the original BOA: the development of a Local Search Algorithm Based on Mutation (LSAM) operator to avoid local optima problem and the use of LSAM to improve BOA solutions diversity. To demonstrate the efficiency and superiority of the proposed DBOA algorithm, 20 benchmark datasets from the UCI repository are employed. The classification accuracy, the fitness values, the number of selected features, the statistical results, and convergence curves are reported for DBOA and its competing algorithms. These results demonstrate that DBOA significantly outperforms the comparative algorithms on the majority of the used performance metrics.

Machine learning predicts putative hematopoietic stem cells within large single-cell transcriptomics data sets

Hematopoietic stem cells (HSCs) are an essential source and reservoir for normal hematopoiesis, and their function is compromised in many blood disorders. HSC research has benefitted from the recent development of single-cell molecular profiling technologies, where single-cell RNA sequencing (scRNA-seq) in particular has rapidly become an established method to profile HSCs and related hematopoietic populations. The classic definition of HSCs relies on transplantation assays, which have been used to validate HSC function for cell populations defined by flow cytometry. Flow cytometry information for single cells, however, is not available for many new high-throughput scRNA-seq methods, thus highlighting an urgent need for the establishment of alternative ways to pinpoint the likely HSCs within large scRNA-seq data sets. To address this, we tested a range of machine learning approaches and developed a tool, hscScore, to score single-cell transcriptomes from murine bone marrow based on their similarity to gene expression profiles of validated HSCs. We evaluated hscScore across scRNA-seq data from different laboratories, which allowed us to establish a robust method that functions across different technologies. To facilitate broad adoption of hscScore by the wider hematopoiesis community, we have made the trained model and example code freely available online. In summary, our method hscScore provides fast identification of mouse bone marrow HSCs from scRNA-seq measurements and represents a broadly useful tool for analysis of single-cell gene expression data.

Bidirectional reservoir networks trained using SVM $$+$$ + privileged information for manufacturing process modeling

In the last decade, a wide range of machine learning approaches were proposed and experimented to model highly nonlinear manufacturing processes. However, improving the performance of such models is challenging due to the complexity and high dimensionality of the manufacturing processes in general. In this paper, we propose bidirectional echo state reservoir networks (Bi-ESNs) trained using support vector machine privileged information method (SVM\(+\)) to model a winding machine process. The proposed model will be applied, tested and compared to reported models in the literature such as classical ESN with linear regression, ESN with a linear SVM readout, genetic programming, feedfoward neural network with backpropagation, radial basis function network, adaptive neural fuzzy inference system and local linear wavelet neural network. The developed results show that Bi-ESNs trained with SVM\(+\) are promising. It was able to provide better generalization performance compared to other models.

On the Encoding of Proteins for Disordered Regions Prediction

Disordered regions, i.e., regions of proteins that do not adopt a stable three-dimensional structure, have been shown to play various and critical roles in many biological processes. Predicting and understanding their formation is therefore a key sub-problem of protein structure and function inference. A wide range of machine learning approaches have been developed to automatically predict disordered regions of proteins. One key factor of the success of these methods is the way in which protein information is encoded into features. Recently, we have proposed a systematic methodology to study the relevance of various feature encodings in the context of disulfide connectivity pattern prediction. In the present paper, we adapt this methodology to the problem of predicting disordered regions and assess it on proteins from the 10th CASP competition, as well as on a very large subset of proteins extracted from PDB. Our results, obtained with ensembles of extremely randomized trees, highlight a novel feature function encoding the proximity of residues according to their accessibility to the solvent, which is playing the second most important role in the prediction of disordered regions, just after evolutionary information. Furthermore, even though our approach treats each residue independently, our results are very competitive in terms of accuracy with respect to the state-of-the-art. A web-application is available at http://m24.giga.ulg.ac.be:81/x3Disorder.

Extracting knowledge from biomedical data through Logic Learning Machines and Rulex

Extracting knowledge from biomedical data through Logic Learning Machines and Rulex