Conventional Machine Research Articles

A Stochastic Regression and Sentiment Analysis-Based Approach for Forecasting Trends in the Stock Market

Stochastic regression problems especially applied to time series forecasting problems often encounter the challenge of volatility and unpredictable seasonality in datasets. One such application happens to forecasting the movement of stock markets globally. Stock data across the world exhibits an intrinsic baseline noise, which often can’t be correlated to previous temporal patterns. Rather, such divergences are an effect of multiple non-numeric variables which govern stock movement such as political scenarios, trade wars, imminent economic waxing and waning trends to name a few. However, it is essential to incorporate these factors while training a model so as to design a strongly sentient and anticipatory regression algorithm which is both pervasive and robust, encompassing as many numeric and non-numeric factors as possible. While conventional machine learning and deep learning approaches tend to leverage tested model architectures, they often miss out on the optimization of both training data and training method to yield potentially high forecasting accuracy. The paper combines stochastic regression in terms of momentum based training and data optimization which is shown to exhibit considerable lower forecasting error compared to existing approaches. To encompass a wider feature space, sentiment analysis has been incorporated to correlate stock movements with public opinions. The proposed approach has been tested on a multitude of benchmark datasets, to validate the performance of the approach. The mean absolute percentage error (MAPE) and regression values have been selected as primary metrics for the performance evaluation. The proposed approach attains a mean MAPE value of 3.22%. The analysis of the MAPE and forecasting accuracy w.r.t. existing benchmark approaches proves the improved forecasting performance of the proposed approach over a period of 300 days.

Time-series data clustering with load-shape preservation for identifying residential energy consumption behaviors

Categorizing residential energy demand patterns is a principal task for demand-side management (DSM) and energy-saving strategies. While deep learning (DL)-based clustering offers a promising alternative to conventional machine learning (ML), DL’s advantages and disadvantages over ML still remain unclear in identifying energy demand patterns. Moreover, prevalent DL-based clustering can suffer from catastrophic feature distortion when capturing load-shape information from energy-load data, leading to erroneous pattern identification. To address these issues, we propose integrating a load-shape preservation mechanism into representative DL-based clustering and investigate its effectiveness in categorizing energy demand patterns, compared to existing ML and DL. We experiment and compare the three clustering approaches using one-year residential energy-load data. Results show that the proposed DL, equipped with load-shape preservation, outperformed ML quantitatively and closely aligns with the baseline DL’s performance. This is particularly significant considering that the baseline DL prioritizes quantitative enhancements, sometimes compromising load-shape precision. Furthermore, the proposed DL discovered more diverse energy demand patterns than the baseline ML and DL, while producing more human-agreeable results. This finding underscores the pivotal role of load-shape preservation in enhancing data clustering and demand pattern recognition in the real-world. These benefits will facilitate personalized DSM interventions and foster residents’ energy-saving behaviors.

Exploration and demonstration of explainable machine learning models in prosthetic rehabilitation-based gait analysis.

Quantitative gait analysis is important for understanding the non-typical walking patterns associated with mobility impairments. Conventional linear statistical methods and machine learning (ML) models are commonly used to assess gait performance and related changes in the gait parameters. Nonetheless, explainable machine learning provides an alternative technique for distinguishing the significant and influential gait changes stemming from a given intervention. The goal of this work was to demonstrate the use of explainable ML models in gait analysis for prosthetic rehabilitation in both population- and sample-based interpretability analyses. Models were developed to classify amputee gait with two types of prosthetic knee joints. Sagittal plane gait patterns of 21 individuals with unilateral transfemoral amputations were video-recorded and 19 spatiotemporal and kinematic gait parameters were extracted and included in the models. Four ML models-logistic regression, support vector machine, random forest, and LightGBM-were assessed and tested for accuracy and precision. The Shapley Additive exPlanations (SHAP) framework was applied to examine global and local interpretability. Random Forest yielded the highest classification accuracy (98.3%). The SHAP framework quantified the level of influence of each gait parameter in the models where knee flexion-related parameters were found the most influential factors in yielding the outcomes of the models. The sample-based explainable ML provided additional insights over the population-based analyses, including an understanding of the effect of the knee type on the walking style of a specific sample, and whether or not it agreed with global interpretations. It was concluded that explainable ML models can be powerful tools for the assessment of gait-related clinical interventions, revealing important parameters that may be overlooked using conventional statistical methods.

Comparative analysis of conventional and ensemble machine learning models for predicting split tensile strength in thermal stressed SCM-blended lightweight concrete

This study aimedto predict the split tensile strength (STS) of lightweight concrete having silica fume (SF) as supplementary cementitious material (SCM) under thermal stresses. Machine learning (ML) models including a conventional machine learning (CML) model as Artificial Neural Networks (ANN) and ensemble machine learning (EML) models, particularly Random Forest (RF) and Gradient Boosting (GB) were developed to predict the STS. Dataset comprising of 96 instances was splitted into 80% for training and 20% for testing phase respectively. With the help of the 80% training data and subsequent hyperparameter tuning, various CML and EML models were developed. ML models were analysed statistically for both the training and testing phases, under the model performanceparameters, line diagrams, and Taylor diagrams. After the thorough statistical analysis, it was concluded that the GB as an EML model outperformed all other ML models. This was demonstrated by the fact that it had constantly higher model performance parameter values of RMSE, MAE and R2 as 0.151, 0.115 and 0.977 for training and 0.23, 0.196 and 0.908 for testing respectively. Also, from the line diagrams which shows a good correlation with the actual and predicted STS values and from the Taylor diagram of the various ML models for both the phases, it was concluded that the GB model is found superior model for this specific study.

An Ultralow-Power Real-Time Machine Learning Based fNIRS Motion Artifacts Detection.

Due to iterative matrix multiplications or gradient computations, machine learning modules often require a large amount of processing power and memory. As a result, they are often not feasible for use in wearable devices, which have limited processing power and memory. In this study, we propose an ultralow-power and real-time machine learning-based motion artifact detection module for functional near-infrared spectroscopy (fNIRS) systems. We achieved a high classification accuracy of 97.42%, low field-programmable gate array (FPGA) resource utilization of 38354 lookup tables and 6024 flip-flops, as well as low power consumption of 0.021 W in dynamic power. These results outperform conventional CPU support vector machine (SVM) methods and other state-of-the-art SVM implementations. This study has demonstrated that an FPGA-based fNIRS motion artifact classifier can be exploited while meeting low power and resource constraints, which are crucial in embedded hardware systems while keeping high classification accuracy.

Real-time classification of tumour and non-tumour tissue in colorectal cancer using diffuse reflectance spectroscopy and neural networks to aid margin assessment.

Colorectal cancer is the third most commonly diagnosed malignancy and the second leading cause of mortality worldwide. A positive resection margin following surgery for colorectal cancer is linked with higher rates of local recurrence and poorer survival. We investigated diffuse reflectance spectroscopy (DRS) to distinguish tumour and non-tumour tissue in ex vivo colorectal specimens, to aid margin assessment and provide augmented visual maps to the surgeon in real-time. Patients undergoing elective colorectal cancer resection surgery at a London-based hospital were prospectively recruited. A hand-held DRS probe was used on the surface of freshly resected ex vivo colorectal tissue. Spectral data was acquired for tumour and non-tumour tissue. Binary classification was achieved using conventional machine learning classifiers and a convolutional neural network (CNN), which were evaluated in terms of sensitivity, specificity, accuracy and the area under the curve. A total of 7692 mean spectra were obtained for tumour and non-tumour colorectal tissue. The CNN-based classifier was the best performing machine learning algorithm, when compared to contrastive approaches, for differentiating tumour and non-tumour colorectal tissue, with an overall diagnostic accuracy of 90.8% and area under the curve of 96.8%. Live on-screen classification of tissue type was achieved using a graduated colourmap. A high diagnostic accuracy for a DRS probe and tracking system to differentiate ex vivo tumour and non-tumour colorectal tissue in real-time with on-screen visual feedback was highlighted by this study. Further in vivo studies are needed to ensure integration into a surgical workflow.

Critical Review on Water Quality Analysis using IoT and Machine Learning Models

Water quality and its management are the most precise concerns confronting humanity globally. This article evaluates the various sensors used for water quality monitoring and focuses on the water quality index considering the multiple physical, chemical, and biological parameters. A Review of Internet of Things (IoT) research for water quality monitoring and analysis, sensors used for water quality can help remote monitoring of the water quality parameters using various IoT-based sensors that convey the assembled estimations utilizing Low-Power Wide Area Network innovations. Overall, the IoT system was 95% accurate in measuring pH, Turbidity, TDS, and Temperature, while the traditional method was only 85% accurate. Also, this study reviewed the different A.I. techniques used to assess water quality, including conventional machine learning techniques, Support Vector Machines, Deep Neural Networks, and K-nearest neighbors. Compared to traditional methods, machine learning and deep learning can significantly increase the accuracy of measurements of groundwater quality. However, various variables, such as the caliber of the training data, the water quality metrics' complexity, and the monitoring frequency, will affect the accuracy. The geographical information system (GIS.) is used for spatial data analysis and managing water resources. The quality of its data is also reviewed in the paper. Based on these analyses, the study has forecasted the future sensors, Geospatial Technology, and machine learning techniques for water quality monitoring and analysis.

Development and validation of a radiomics-based model for predicting osteoporosis in patients with lumbar compression fractures

BACKGROUNDOsteoporosis, a metabolic bone disorder, markedly elevates fracture risks, with vertebral compression fractures being predominant. Antiosteoporotic treatments for patients with osteoporotic vertebral compression fractures (OVCF) lessen both the occurrence of subsequent fractures and associated pain. Thus, diagnosing osteoporosis in OVCF patients is vital. PURPOSEThe aim of this study was to develop a predictive radiographic model using T1 sequence MRI images to accurately determine whether patients with lumbar spine compression fractures also have osteoporosis. STUDY DESIGNRetrospective cohort study. PATIENT SAMPLEPatients over 45 years of age diagnosed with a fresh lumbar compression fracture. OUTCOME MEASURESDiagnostic accuracy of the model (area under the ROC curve). METHODSThe study retrospectively collected clinical and imaging data (MRI and DEXA) from hospitalized lumbar compression fracture patients (L1–L4) aged 45 years or older between January 2021 and June 2023. Using the pyradiomics package in Python, features from the lumbar compression fracture vertebral region of interest (ROI) were extracted. Downscaling of the extracted features was performed using the Mann-Whitney U test and the least absolute shrinkage selection operator (LASSO) algorithm. Subsequently, six machine learning models (Naive Bayes, Support Vector Machine [SVM], Decision Tree, Random Forest, Extreme Gradient Boosting [XGBoost], and Light Gradient Boosting Machine [LightGBM]) were employed to train and validate these features in predicting osteoporosis comorbidity in OVCF patients. RESULTSA total of 128 participants, 79 in the osteoporotic group and 49 in the non-osteoporotic group, met the study's inclusion and exclusion criteria. From the T1 sequence MRI images, 1906 imaging features were extracted in both groups. Utilizing the Mann-Whitney U test, 365 radiologic features were selected out of the initial 1,906. Ultimately, the lasso algorithm identified 14 significant radiological features. These features, incorporated into six conventional machine learning algorithms, demonstrated successful prediction of osteoporosis in the validation set. The NaiveBayes model yielded an area under the receiver operating characteristic curve (AUC) of 0.84, sensitivity of 0.87, specificity of 0.70, and accuracy of 0.81. CONCLUSIONSA NaiveBayes machine learning algorithm can predict osteoporosis in OVCF patients using t1-sequence MRI images of lumbar compression fractures. This approach aims to obviate the necessity for further osteoporosis assessments, diminish patient exposure to radiation, and bolster the clinical care of patients with OVCF.

A Weighted Stacking Ensemble Model With Sampling for Fake Reviews Detection

Customers use reviews as a primary source of information to judge a product or service. Positive reviews help boost companies’ reputations, increasing their revenue by attracting new clients, and increasing the purchasing order size. On the other hand, negative reviews significantly reduce sales, which might be the case due to competitive advantage. Organizations can use fake (i.e., misleading or fraudulent) reviews to generate fast profits by deceiving customers into buying their products. Recently, various methods to assess the legitimacy of reviews have been introduced using advances in machine learning. However, existing methods fall short of achieving highly accurate detection results for unbalanced classes. We aimed to create a spam review identification model using ensemble-based learning while balancing classes using sampling techniques. This article proposes a weighted stacking ensemble model with sampling (WSEM-S) for efficient fake reviews detection. We used <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$n$</tex-math> </inline-formula> -gram models to effectively model language data for feature retrieval. The experimental results on three customer reviews datasets: YELPNYC, Deceptive Opinion Spam Corpus (DOSC) v1.4, and Deception datasets show that the proposed model outperforms the conventional machine learning techniques Naïve Bayes, logistic regression, K-nearest neighbor (KNN), random forest, extreme gradient boosting (XGBoost), and convolutional neural network (CNN) as well as the state-of-the-art ensemble models.

An Empirical Evaluation of Adapting Hybrid Parameters for CNN-based Sentiment Analysis

Sentiment analysis aims to understand human emotions and perceptions through various machine-learning pipelines. However, feature engineering and inherent semantic gap constraints often hinder conventional machine learning techniques and limit their accuracy. Newer neural network models have been proposed to automate the feature learning process and enrich learned features with word contextual embeddings to identify their semantic orientations to address these challenges. This article aims to analyze the influence of different factors on the accuracy of sentiment classification predictions by employing Feedforward and Convolutional Neural Networks. To assess the performance of these neural network models, we utilize four diverse real-world datasets, namely 50,000 movie reviews from IMDB, 10,662 sentences from LightSide Movie_Reviews, 300 public movie reviews, and 1,600,000 tweets extracted from Sentiment140. We experimentally investigate the impact of exploiting GloVe word embeddings on enriching feature vectors extracted from sentiment sentences. Findings indicate that using larger dimensions of GloVe word embeddings increases the sentiment classification accuracy. In particular, results demonstrate that the accuracy of the CNN with a larger feature map, a smaller filter window, and the ReLU activation function in the convolutional layer was 90.56% using the IMDB dataset. In comparison, it was 80.73% and 77.64% using the sentiment140 and the 300 sentiment sentences dataset, respectively. However, it is worth mentioning that, with large-size sentiment sentences (LightSide’s Movie Reviews) and using the same parameters, only a 64.44% level of accuracy was achieved.

Sentiment analysis with hotel customer reviews using FNet

Recent research has focused on opinion mining from public sentiments using natural language processing (NLP) and machine learning (ML) techniques. Transformer-based models, such as bidirectional encoder representations from transformers (BERT), excel in extracting semantic information but are resourceintensive. Google’s new research, mixing tokens with fourier transform, also known as FNet, replaced BERT’s attention mechanism with a non-parameterized fourier transform, aiming to reduce training time without compromising performance. This study fine-tuned the FNet model with a publicly available Kaggle hotel review dataset and investigated the performance of this dataset in both FNet and BERT architectures along with conventional machine learning models such as long short-term memory (LSTM) and support vector machine (SVM). Results revealed that FNet significantly reduces the training time by almost 20% and memory utilization by nearly 60% compared to BERT. The highest test accuracy observed in this experiment by FNet was 80.27% which is nearly 97.85% of BERT’s performance with identical parameters.

Rockburst classification based on cross reconstruction learning under small-sample condition

Rockburst is a prevalent geological hazard in deep geotechnical engineering, and its accurate prognostication is vital for prevention measures. Consequently, this research proffers a pioneering classification prediction methodology, namely Cross Reconstruction Learning (CR), underpinned by conventional machine learning algorithms and metric learning strategies. Initially, this technique partitions and restructures the original dataset, where each sample feature intersects and reconfigures with features from other samples within the set. During this amalgamation, new samples are assigned labels based on the degree of divergence or congruity between two sets of sample labels, thereby forming a new set of samples. Subsequently, an array of machine learning algorithms is utilized to train and test this new dataset. Ultimately, employing a universal class voting mechanism and decoding test set results through probability assignment, the predicted labels are converted back into rock burst outcomes, thereby determining the final prediction classification. The proposed model was trained on a database encompassing 239 instance samples, and its performance was validated against the currently proficient models (KNN, XGBoost, and Random Forest algorithms) employed in rock burst prediction. The outcome revealed a decline in the performance metrics of all three machine learning algorithms when interfaced with the Cross Reconstruction learning method, particularly the KNN algorithm, owing to the doubled feature dimensions in the combined dataset. However, the metrics of ensemble models, XGBoost and Random Forest, exhibited a notable improvement compared to the original classification models. On comparing multiple performance metrics, it was discovered that the CR-XGBoost model outperformed others across all evaluations, thereby offering significant guidance for practical engineering applications.

Comparative study of machine learning models for predicting the compressive strength of concrete using Non-Destructive Testing methods

This study proposes and compares the conventional machine learning (CML) models as artificial neural networks (ANN), support vector machines (SVM) and ensemble machine learning (EML) models namely, random forest (RF), and gradient boosting (GB). These four-machine learning (ML) models were used to predict the compressive strength (CS) of concrete based on two non-destructive testing (NDT) methods, namely ultrasonic pulse velocity (UPV) and rebound hammer number (RHN). The article uses a large and reliable database of experimental data, collected from the literature to train and test the models, and evaluates their performance using various metrics, scatter plots and a Taylor diagram. It was found that gradient boosting (GB) is the most accurate and robust model for this study, followed by RF model, while ANN and SVM have lower performance. For this study, it suggests that EML models are more suitable and reliable than CML models for predicting the CS of concrete based on NDT methods.

Comparison of Adam Optimization and RMS prop in Minangkabau-Indonesian Bidirectional Translation with Neural Machine Translation

Language is a tool humans use to establish communication. Still, the language used is one language and between regions or nations with their languages. Indonesia is a country that has a diversity of second languages and is the fourth most populous country in the world. It is recorded that Indonesia has nearly 800 regional languages, but research activities in natural language processing are still lacking. Minangkabau is an endangered language spoken by the Minangkabau people in Indonesia's West Sumatra province. According to UNESCO, the Minangkabau language is listed as a language that is "definitely endangered," with only around 5 million speakers worldwide. This study uses neural machine translation (NMT) to create a formula based on this information. Neural machine translation, in contrast to conventional statistical machine translation, intends to build a single neural network that can be built up to achieve the best performance. Because it can simultaneously hold memory for a long time, comprehend complicated relationships in data, and provide information that is very important in determining the outcome of translation, LSTM is one of the most powerful machine-learning techniques for translating languages. The BLUE score is utilized in the NMT evaluation. The test results use 520 Minangkabau sentences, conducting tests based on the number of epochs ranging from 100-1000, resulting in optimization using Adam being better than optimization RMSprop. This is evidenced by the results of the best BLUE-1 score of 0.997816 using 1000 epochs.

Review on VEND IT!- A Smart Vending Machine

Abstract: Traditional vending machines, though reliable, have often lagged behind in terms of convenience, personalization, and sustainability. As society rely on digital connectivity and automation continues to grow, the need for a smarter, more responsive vending solution has become increasingly evident. In a world marked by constant technological innovation and evolving consumer preferences, the time has come to reimagine the traditional vending machine. "Vend It" is a visionary project that aims to reinvent the conventional vending machine, leveraging cutting-edge smart technology to create a seamless, efficient, and user-friendly vending experience. VEND IT! encompasses a comprehensive suite of innovations, ranging from intelligent inventory management to user-centric interface design, and remote monitoring and management capabilities.

Chiller fault diagnosis based on improved variational autoencoder and co-training framework: A case study of insufficient samples

Implementing efficient automatic fault diagnosis is critical for saving energy and minimizing financial losses in the heating ventilation air-conditioning (HVAC) systems of commercial buildings. However, the limited quantity and weak features of fault samples acquired during HVAC operations hinder the effectiveness of conventional machine learning-based fault diagnosis methodologies. This paper proposes a method based on an improved conditional variational autoencoder (MCVAE) and co-training (CT) ideology to address the issue of insufficient training samples. Initially, we employ MCVAE to synthesize an extensive dataset of chiller fault samples from the original training dataset. Subsequently, the beneficial samples for training our fault diagnostic classifier, namely high-quality samples, are selected from the generated dataset using the CT-based framework. Finally, the selected high-quality samples are merged into the original training dataset to train the ultimate fault classifiers. Experimental results demonstrate that our proposed method outperforms in effectiveness and efficiency compared to recently published methods. For instance, in the case of fault level 1 compared to the suboptimal model, our approach exhibits improvements of 2.41% when each type has 5 fault samples.

AI perceives like a local: predicting citizen deprivation perception using satellite imagery

Deprived urban areas, commonly referred to as ‘slums,’ are the consequence of unprecedented urbanisation. Previous studies have highlighted the potential of Artificial Intelligence (AI) and Earth Observation (EO) in capturing physical aspects of urban deprivation. However, little research has explored AI’s ability to predict how locals perceive deprivation. This research aims to develop a method to predict citizens’ perception of deprivation using satellite imagery, citizen science, and AI. A deprivation perception score was computed from slum-citizens’ votes. Then, AI was used to model this score, and results indicate that it can effectively predict perception, with deep learning outperforming conventional machine learning. By leveraging AI and EO, policymakers can comprehend the underlying patterns of urban deprivation, enabling targeted interventions based on citizens’ needs. As over a quarter of the global urban population resides in slums, this tool can help prioritise citizens’ requirements, providing evidence for implementing urban upgrading policies aligned with SDG-11.

A comparative study of the deep learning based model and the conventional machine learning based models in human activity recognition

Human activity recognition (HAR) has been widely studied as a research field in human behavior analysis due to its huge potential in various application domains such as health care and behavioral science. Recently, deep learning (DL) based methods have also been successfully applied to predict various human activities. This research aims at building different Python-based models to perform HAR using smartphones and calculating and comparing the accuracy of the models to select the optimal one. Four models were built to classify and predict human activities: Deep Convolutional Neural Network (DCNN), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF). The results of the experiments in this paper show that the Deep Convolutional Neural Network achieves an average recognition accuracy rate of 95.49%, exceeding the other three models. The underlying reason may be that Deep Convolutional Neural Network is based on a more advanced algorithm deep learning technique.

Deep Learning in Single-cell Analysis

Single-cell technologies are revolutionizing the entire field of biology. The large volumes of data generated by single-cell technologies are high dimensional, sparse, and heterogeneous and have complicated dependency structures, making analyses using conventional machine learning approaches challenging and impractical. In tackling these challenges, deep learning often demonstrates superior performance compared to traditional machine learning methods. In this work, we give a comprehensive survey on deep learning in single-cell analysis. We first introduce background on single-cell technologies and their development, as well as fundamental concepts of deep learning including the most popular deep architectures. We present an overview of the single-cell analytic pipeline pursued in research applications while noting divergences due to data sources or specific applications. We then review seven popular tasks spanning different stages of the single-cell analysis pipeline, including multimodal integration, imputation, clustering, spatial domain identification, cell-type deconvolution, cell segmentation, and cell-type annotation. Under each task, we describe the most recent developments in classical and deep learning methods and discuss their advantages and disadvantages. Deep learning tools and benchmark datasets are also summarized for each task. Finally, we discuss the future directions and the most recent challenges. This survey will serve as a reference for biologists and computer scientists, encouraging collaborations.

Advanced Framework for Multi-Modal Healthcare Data Integration: Leveraging HPC with GPU Computing and CNN Architecture in CDSS

In this study, we shall be looking at the challenges involved in integrating multi-modal healthcare data in the clinical decision support systems (CDSS). We propose the Automated Multi-Modal Data Integration (AMMI-CDSS) algorithm, which will utilize the latest high-performance computing (HPC) techniques such as the Convolutional Neural Network (CNN) architecture and the Graphics Processing Unit (GPU) computing to provide precise and rapid analysis. Which features will be extracted, multi-modal data will be merged, data will be prepared and algorithms developed in a distributed computing environment. We illustrate how AMMI-CDSS through the use of real world datasets such as wearable sensors data, medical imaging, genetic data, and electronic health records (EHRs), can improve the clinical decision support. By performing harmonization of the diverse data sources into a unique dataset after thorough data preprocessing and complex calculations, AMMI-CDSS provides the analysis with better quality and coherence.&#x0D; Our study allow us to make conclusion about how HPC-based CDSS models can be compared to conventional machine learning ones using their scalability and performance as key metrics. We enrich CDSS with the methodical framework for one-by-one testing and evaluation of proposed models and multi-modal healthcare data analysis. Future research might explore novel methods for integrating diverse types of healthcare data, as well as enhancing the HPC-based CDSS models by keeping them up-to-date.