Bidirectional Long Short-term Memory Model Research Articles

Stock prediction using Bi-LSTM and RoBERTa sentiment analysis

Stock price forecasting is a widespread research issue in the field of finance. Stock price prediction can give investors more accurate advice and yield higher returns. With the popularity of social media, more and more people choose to express their opinions, and more and more media outlets published news on social media. This paper mainly conducts stock analysis and prediction by presenting the RoBERTa-BiLSTM model with the Bahdanau attention mechanism. It consists of two existing models: the RoBERTa and bidirectional LSTM models. The experimental results show that this model is more accurate in predicting stock prices than the Multi-Layer Perception (MLP) and LSTM models. Inputs to the model include historical AAPL stock price data and sentiment data analysed by news and Twitter tweet sentiment. In this paper, the high-precision RoBERTa-BiLSTM model for stock prediction was proposed, which provides a solution for stock prediction. The practical applications and the typical problem of most studies about stock prediction called the time-shifted problem, are provided in this paper and also have research value.

Soft sensing of NOx emission from waste incineration process based on data de-noising and bidirectional long short-term memory neural networks.

Continuous emission monitoring system is commonly employed to monitor NOx emissions in municipal solid waste incineration (MSWI) processes. However, it still encounters the challenges of regular maintenance and measurement lag. These issues significantly impact the accurate and stable control of NOx emissions. Therefore, developing a soft NOx emission sensor to complement hardware monitoring becomes imperative. Considering data noise, dynamic nonlinearity, time series characteristics and volatility in the MSWI process, this article introduces a soft sensor model for NOx emission prediction utilizing the complete ensemble empirical mode decomposition adaptive noise (CEEMDAN)-wavelet threshold (WT) method and bidirectional long short-term memory (Bi-LSTM). Firstly, the original data signal is decomposed into a group of intrinsic mode functions (IMFs) using the CEEMDAN. Subsequently, the WT processes the high-frequency IMFs that are noise-dominant. Then, all IMFs are reconstructed to obtain the denoized signal. Finally, the Bi-LSTM model is employed to predict NOx emissions. Compared to conventional modelling approaches, the model proposed in this article demonstrates the best predictive performance. The mean absolute percentage error, root-mean-squared error and average absolute error on the test set of the proposed model are 3.75%, 5.34 mg m-3 and 4.34 mg m-3, respectively. The proposed model provides a new method to soft sensing NOx emissions. It holds significant practical value for precise and stable monitoring of NOx emissions in MSWI processes and provides a reference for research on modelling key process parameters.

Prediction of UCS and BTS under freeze-thaw conditions in the NW himalayan rock mass using petrographic analysis and laboratory testing

Repeated freeze-thaw (F&T) cycles substantially harm the durability of rocks, heightening the potential for landslides, rockslides, and avalanches. The current work investigates the effect of the F&T cycle on rock mass (biotite schist) samples. For this purpose, 32 rock samples were prepared and gathered from eight distinct locations in the northwest Himalayan region. For each sample, petrographical analysis and laboratory testing such as uniaxial compressive strength (UCS) and Brazilian tensile strength (BTS) are investigated at repeated (0th, 10th, 20th, and 30th) F&T cycles. Additionally, machine learning (ML) sequential models such as recurrent neural networks (RNN), gated recurrent units (GRU), and bi-directional long short-term memory (Bi-LSTM) are constructed to estimate the UCS and BTS under F&T conditions. Petrographical results show no change in the mineral indices, while there is a noticeable increase in aspect ratio but a significant decline in mean grain size with each successive 10th cycle, suggesting sample damage. The study also provides a comprehensive assessment of the ML models' performance, highlighting the Bi-LSTM model's superior accuracy among all models in terms of R2 (0.9850) and RMSLE (0.0100) during the TR stage and R2 (0.9020) and RMSLE (0.0170) during the TS stage for UCS prediction. Similarly, BTS prediction also shows superior precision, recording an R2 (0.7543) and RMSLE (0.0345) during TR and R2 (0.7404) and RMSLE (0.0213) during TS stages. The present study also explores the heatmap, line diagram, regression analysis, 2D kernel density plot, Taylor diagram, and DDR criterion for evaluating the model performance more clearly.

Classification Model for Real-Time Monitoring of Machining Status of Turned Workpieces

The occurrence of tool chatter can have a detrimental impact on the quality of the workpiece. In order to improve surface quality, machining stability, and reduce tool wear cycles, it is essential to monitor the workpiece machining process in real time during the turning process. This paper presents a tool chatter state recognition model based on a denoising autoencoder (DAE) for feature dimensionality reduction and a bidirectional long short-term memory (BiLSTM) network. This study examines the feature dimensionality reduction method of the DAE, whereby the reduced-dimensional data are concatenated and input into the BiLSTM model for training. This approach reduces the learning difficulty of the network and enhances its anti-interference capability. Turning experiments were conducted on a SK50P lathe to collect the dataset for model performance validation. The experimental results and analysis indicate that the proposed DAE-BiLSTM model outperforms other models in terms of prediction and classification accuracy in distinguishing between stable machining, over-machining, and severe chatter stages in turning chatter state recognition. The overall classification accuracy reached 96.28%.

Wind speed multi-step prediction based on the comparison of wind characteristics and error correction: Focusing on periodic thermally-developed winds

To leverage the abundant wind energy resources available in mountainous regions, an increasing number of wind power facilities and associated transportation infrastructure are being constructed in these areas. There are different types of intense winds in mountainous areas and the non-linear characteristics of the different intense winds vary significantly. The periodic thermally-developed winds are the commonly occurring type of intense wind. Based on the comparison of wind characteristics, this research proposes a prediction model for the periodic thermally-developed winds. The research utilizes the Bi-directional Long Short-Term Memory model(Bi-LSTM), which has been optimized by the Bayesian optimization(BO), to predict wind speed. Based on the relationship between wind field characteristics in periodic thermally-developed winds, a regression correction module is proposed using a joint probability model. Additionally, to address the volatility and uncertainty of wind speed, a Gaussian mixture model is employed for probability prediction. The experimental results demonstrate several findings: 1. The BO-Bi-LSTM model accurately and reliably predicts multi-step results for periodic thermally-developed winds in mountainous areas. 2. The proposed correction module, when combined with other models, yields improved wind speed accuracy compared to the uncorrected values. 3. The Gaussian mixture model produces probabilistic prediction results that consider both interval coverage and interval bandwidth simultaneously. 4. The proposed model can successfully predict periodic thermally-developed winds in mountainous areas for practical engineering purposes, resulting in reliable probabilistic prediction outcomes. It is of great significance for the utilization of wind energy resources and the driving safety of vehicles on ancillary structures.

Parsimonious airfoil Parameterisation: A deep learning framework with Bidirectional LSTM and Gaussian Mixture models

The choice of airfoil parameterisation method significantly influences the overall wing optimisation performance by affecting the flexibility and computational efficiency of the process. Ideally, one should be able to intuitively constrain airfoil shape and structural characteristics as input to the optimisation process. Current parameterisation techniques lack the flexibility to generate airfoils efficiently by specifying parsimonious shape and structural features. To address this limitation, a deep learning framework is proposed, enabling conditional airfoil generation from an airfoil’s shape and structural feature definition.Specifically, we demonstrate the application of Bidirectional Long Short Term Memory models and Bayesian Gaussian Mixture models to derive airfoil coordinates from a compact set of shape and structural characteristics that we define. The proposed framework is shown to achieve favorable airfoil performance optimisation due to improved exploration and exploitation of the design space, compared to traditional approaches. Overall, the proposed optimisation framework is able to realise a 9.04% performance improvement over an airfoil design optimised with traditional parameterisation techniques.

A Photovoltaic Prediction Model with Integrated Attention Mechanism

Solar energy has become a promising renewable energy source, offering significant opportunities for photovoltaic (PV) systems. Accurate and reliable PV generation forecasts are crucial for efficient grid integration and optimized system planning. However, the complexity of environmental factors, including seasonal and daily patterns, as well as social behaviors and user habits, presents significant challenges. Traditional prediction models often struggle with capturing the complex nonlinear dynamics in multivariate time series, leading to low prediction accuracy. To address this issue, this paper proposes a new PV power prediction method that considers factors such as light, air pressure, wind direction, and social behavior, assigning different weights to them to accurately extract nonlinear feature relationships. The framework integrates long short-term memory (LSTM) and gated recurrent units (GRU) to capture local time features, while bidirectional LSTM (BiLSTM) and an attention mechanism extract global spatiotemporal relationships, effectively capturing key features related to historical output. This improves the accuracy of multi-step predictions. To verify the feasibility of the method for multivariate time series, we conducted experiments using PV power prediction as a scenario and compared the results with LSTM, CNN, BiLSTM, CNN-LSTM and GRU models. The experimental results show that the proposed method outperforms these models, with a mean absolute error (MAE) of 12.133, root mean square error (RMSE) of 14.234, mean absolute percentage error (MAPE) of 2.1%, and a coefficient of determination (R2) of 0.895. These results indicate the effectiveness and potential of the method in PV prediction tasks.

Predicting abrupt depletion of dissolved oxygen in Chaohu lake using CNN-BiLSTM with improved attention mechanism

Depletion of dissolved oxygen (DO) is a significant incentive for biological catastrophic events in freshwater lakes. Although predicting the DO concentrations in lakes with high-frequency real-time data to prevent hypoxic events is effective, few related experimental studies were made. In this study, a short-term predicting model was developed for DO concentrations in three problematic areas in China's Chaohu Lake. To predict the DO concentrations at these representative sites, which coincide with biological abnormal death areas, water quality indicators at the three sampling sites and hydrometeorological features were adopted as input variables. The monitoring data were collected every 4 h between 2020 and 2023 and applied separately to train and test the model at a ratio of 8:2. A new AC-BiLSTM coupling model of the convolution neural network (CNN) and the bidirectional long short-term memory (BiLSTM) with the attention mechanism (AM) was proposed to tackle characteristics of discontinuous dynamic change of DO concentrations in long time series. Compared with the BiLSTM and CNN-BiLSTM models, the AC-BiLSTM showed better performance in the evaluation criteria of MSE, MAE, and R2 and a stronger ability to capture global dependency relationships. Although the prediction accuracy of hypoxic events was slightly worse, the general time series characteristics of abrupt DO depletion were captured. Water temperature regularly affects DO concentrations due to its periodic variations. The high correlation and the universal importance of total nitrogen (TN) and total phosphorus (TP) with DO reveals that point source pollution are critical cause of DO depletion in the freshwater lake. The importance of NTU at the Zhong Miao Station indicates the self-purification capacity of the lake is affected by the flow rate changes brought by the tributaries. Calculating linear correlations of variables in conjunction with a permutation variable importance analysis enhanced the interpretability of the proposed model results. This study demonstrates that the AC-BiLSTM model can complete the task of short-term prediction of DO concentration of lakes and reveal its response features of timing and magnitude of abrupt DO depletion.

Subject-Independent Model for Reconstructing Electrocardiography Signals from Photoplethysmography Signals

Electrocardiography (ECG) is the gold standard for monitoring vital signs and for diagnosing, controlling, and preventing cardiovascular diseases (CVDs). However, ECG requires continuous user participation, and cannot be used for continuous cardiac monitoring. In contrast to ECG, photoplethysmography (PPG) devices do not require continued user involvement, and can offer ongoing and long-term detection capabilities. However, from a medical perspective, ECG can provide more information about the heart. Currently, most existing work contains different signals recorded from the same subject in training and test sets. This study proposes a neural network model based on a 1D convolutional neural network (CNN) and a bidirectional long short-term memory (BiLSTM) network. This neural network model can directly reconstruct ECG signals from PPG signals. The learned features are captured from the CNN model and fed into the BiLSTM model. In order to verify the validity of the model, it is evaluated using the MIMIC II dataset in the completely subject-independent model (records are placed in a training set, and a test set appears once, but the test signal belongs to a record that is not in the training set). The Pearson’s correlation coefficient between the reconstructed ECG and the reference ECG of the proposed model is 0.963 in the completely subject-independence model. The results of the proposed model are better than those of several cited state-of-the-art models. The results of our trained model indicate that we can obtain reconstructed ECGs that are highly similar to reference ECGs in the completely subject-independent model.

A deep learning model employing Bi-LSTM architecture to predict Martian ionosphere electron density using data from the Mars Global Surveyor mission

Planetary scientists and space agencies are highly intrigued by the Martian ionosphere due to its significant impact on upcoming Mars missions. Given its potential influence on satellite communication and navigation systems, measuring the altitude distribution of electron density (ED) values in the Martian ionosphere becomes a vital parameter to consider. The parameter of electron density in the Martian ionosphere serves as a tool to represent and predict the impacts of severe space weather events and solar activity on the Martian ionosphere. This study presents the development of a Bi-directional Long Short-Term Memory (Bi-LSTM) model, a deep machine learning algorithm to predict the Mars ionospheric ED values. To train and evaluate the model’s performance, we employ data samples from the Mars Global Surveyor (MGS) mission. For the preparation of input variables in the Bi-LSTM model, approximately 5600 ED altitude distribution of electron density profiles logged by the MGS between the years 1998 to 2005, which corresponded to the major period of the 23rd solar cycle, are taken into consideration. The performance of the proposed model in predicting both quiet and disturbed times is evaluated using the Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) metrics. The significant minimum values of MAE (0.56 × 109) and RMSE (0.67 × 109) for the quiet day considered and MAE (0.27 × 109) and RMSE (0.31 × 109) for the solar flare disturbed day considered indicate the fair performance of the Bi-LSTM based prediction model. This research’s findings are vital for upcoming Mars missions, as precise predictions of ED of Mars ionosphere are crucial for reliable navigation and communication systems.

Translating strain to stress: a single-layer Bi-LSTM approach to predicting stress-strain curves in alloys during hot deformation

This study introduces a novel deep learning network that integrates a single-layer bidirectional long short-term memory (Bi-LSTM) network with a coding layer to analyze the hot deformation behavior of various alloys. The single-layer Bi-LSTM model adeptly predicts experimental stress–strain curves obtained under different deformation temperatures and strain rates, demonstrating superior effectiveness and excellent performance in modeling hot deformation behaviors of the FGH98 nickel-based alloy and TiAl intermetallic alloy. The present model achieves the coefficient of determination of 0.9051 for FGH98 and 0.9307 for TiAl alloys, whereas the corresponding values of 0.8105 and 0.8356 are obtained by the conventional strain-compensated Sellars constitutive equation (SCS model). Additionally, the mean absolute percentage error of the single-layer Bi-LSTM model are 11.37% for FGH98 and 7.16% for TiAl alloys, while the SCS model gains the corresponding error of 15.29% and 17.01%. These results show that the present model has enhances the predictive accuracy exceeding 10% for both FGH98 and TiAl alloys over the SCS model. Consequently, the proposed single-layer Bi-LSTM model provides substantial potential for optimizing manufacturing processes and improving material properties.

Auto-Diacritization and Stylistic Realization of Arabic Text using Deep Neural Networks

This paper presents QRDiaRec, an advanced diacritization system for Arabic Quranic texts. In Arabic, linguistic style refers to the variations in diacritic markings used to convey different pronunciations, dialects, meanings, and contextual understandings.QRDiaRec addresses the challenge of interpreting Arabic diacritics across multiple linguistic styles, which is crucial for accurate language processing. Unlike traditional systems that generate only one correct form of diacritics, QRDiaRec can recognize and produce multiple valid diacritic forms. This capability is due to its training on a dataset that encompasses seven Quranic linguistic styles.The Qur'an is an ideal case study because it is one text with multiple linguistic patterns, allowing us to recognize different forms of correct diacritization. QRDiaRec employs bidirectional LSTM, GRU, and transformer-based models to convert non-diacritic texts into annotated formats, achieving up to 94.2%accuracy. The system enhances Arabic language processing, impacting NLP, machine translation, and Arabic linguistics.

Forecasting foreign exchange rate volatility using deep learning: Case of US dollar/Algerian dinar during the COVID-19 pandemic

This study explores the application of deep learning techniques in forecasting foreign exchange rate volatility, leveraging the capabilities of neural networks to capture complex patterns and nonlinear relationships within financial data. We applied the auto regressive integrated moving average (ARIMA) and machine learning linear regression (LR) model, deep learning models ( recurrent neural networks (RNN), bidirectional LSTM (Bi-LSTM), long short-term memory (LSTM) and gated recurrent unit (GRU). In terms of forecasting errors, and Python routines were used for such purpose. Morever, In order to investigate the quality of the models used, we compared the performances of these algorithms in US dollar/algerian dinar exchange rate forecasting througt the application of significance satistical tests (R-squared, MSE, RMSE, MAE, MAPE)The results clearly depict that contemporary techniques have been shown to produce more accurate results than conventional regression-based modeling. The machine learning linear regression (LR) model provides the maximum accuracy rate of (99.83%) followed by the RNN models with GRU model (92.27%) , Bi-LSTM model (87.34%), LSTM model (74.68%) and ARIMA model (32.29%) .

Improving clinical abbreviation sense disambiguation using attention-based Bi-LSTM and hybrid balancing techniques in imbalanced datasets.

Clinical abbreviations pose a challenge for clinical decision support systems due to their ambiguity. Additionally, clinical datasets often suffer from class imbalance, hindering the classification of such data. This imbalance leads to classifiers with low accuracy and high error rates. Traditional feature-engineered models struggle with this task, and class imbalance is a known factor that reduces the performance of neural network techniques. This study proposes an attention-based bidirectional long short-term memory (Bi-LSTM) model to improve clinical abbreviation disambiguation in clinical documents. We aim to address the challenges of limited training data and class imbalance by employing data generation techniques like reverse substitution and data augmentation with synonym substitution. We utilise a Bi-LSTM classification model with an attention mechanism to disambiguate each abbreviation. The model's performance is evaluated based on accuracy for each abbreviation. To address the limitations of imbalanced data, we employ data generation techniques to create a more balanced dataset. The evaluation results demonstrate that our data balancing technique significantly improves the model's accuracy by 2.08%. Furthermore, the proposed attention-based Bi-LSTM model achieves an accuracy of 96.09% on the UMN dataset, outperforming state-of-the-art results. Deep neural network methods, particularly Bi-LSTM, offer promising alternatives to traditional feature-engineered models for clinical abbreviation disambiguation. By employing data generation techniques, we can address the challenges posed by limited-resource and imbalanced clinical datasets. This approach leads to a significant improvement in model accuracy for clinical abbreviation disambiguation tasks.

A hybrid deep approach to recognizing student activity and monitoring health physique based on accelerometer data from smartphones

Smartphone sensors have gained considerable traction in Human Activity Recognition (HAR), drawing attention for their diverse applications. Accelerometer data monitoring holds promise in understanding students’ physical activities, fostering healthier lifestyles. This technology tracks exercise routines, sedentary behavior, and overall fitness levels, potentially encouraging better habits, preempting health issues, and bolstering students’ well-being. Traditionally, HAR involved analyzing signals linked to physical activities using handcrafted features. However, recent years have witnessed the integration of deep learning into HAR tasks, leveraging digital physiological signals from smartwatches and learning features automatically from raw sensory data. The Long Short-Term Memory (LSTM) network stands out as a potent algorithm for analyzing physiological signals, promising improved accuracy and scalability in automated signal analysis. In this article, we propose a feature analysis framework for recognizing student activity and monitoring health based on smartphone accelerometer data through an edge computing platform. Our objective is to boost HAR performance by accounting for the dynamic nature of human behavior. Nonetheless, the current LSTM network’s presetting of hidden units and initial learning rate relies on prior knowledge, potentially leading to suboptimal states. To counter this, we employ Bidirectional LSTM (BiLSTM), enhancing sequence processing models. Furthermore, Bayesian optimization aids in fine-tuning the BiLSTM model architecture. Through fivefold cross-validation on training and testing datasets, our model showcases a classification accuracy of 97.5% on the tested dataset. Moreover, edge computing offers real-time processing, reduced latency, enhanced privacy, bandwidth efficiency, offline capabilities, energy efficiency, personalization, and scalability. Extensive experimental results validate that our proposed approach surpasses state-of-the-art methodologies in recognizing human activities and monitoring health based on smartphone accelerometer data.

Judgment prediction from legal documents using Texas wolf optimization based deep BiLSTM model

The complicated nature of legal texts, a lack of labeled data, concerns about fairness, and difficulties with interpretation represent some of the challenges that judicial judgment prediction models encounter. The approach we propose seeks to conquer these challenges by using advanced techniques for deep learning, such as deep Bidirectional Long Short-Term Memory (BiLSTM) networks to recognize complex linguistic patterns and transfer learning to make more efficient use of data. Employing a deep BiLSTM classifier (TWO-BiLSTM) model based on Texas wolf optimization, the research aims to predict legal judgments. To prepare it for evaluation, it initially collects and preprocesses judicial data. Feature extraction involves statistical and Principal component Analysis (PCA) techniques to generate an extensive feature set. The model undergoes training utilizing these features in addition to preprocessed data. A hybrid Texas wolf optimization tactic, based on the optimization of gray wolves and Harris hawks, is employed to boost performance. The ability of the model to accurately and effectively predict legal judgment has been demonstrated by testing it on different sets of judicial data. The model achieved reasonably well in TP 90, having an accuracy of 97.00%. It also achieved exceedingly well in f-score, precision, and recall, having scores of 97.29, 97.10, and 97.19, correspondingly. The model’s effectiveness was further demonstrated in the k-fold 10 assessment, which exhibited 96.00% accuracy and robustness. In addition, using f-score, precision, and recall metrics of 96.25, 96.89, and 95.96, respectively, the model showed outstanding performance. These outstanding results demonstrate the model’s effectiveness and dependability for providing accurate predictions.

Improving accuracy of code smells detection using machine learning with data balancing techniques

Code smells indicate potential symptoms or problems in software due to inefficient design or incomplete implementation. These problems can affect software quality in the long-term. Code smell detection is fundamental to improving software quality and maintainability, reducing software failure risk, and helping to refactor the code. Previous works have applied several prediction methods for code smell detection. However, many of them show that machine learning (ML) and deep learning (DL) techniques are not always suitable for code smell detection due to the problem of imbalanced data. So, data imbalance is the main challenge for ML and DL techniques in detecting code smells. To overcome these challenges, this study aims to present a method for detecting code smell based on DL algorithms (Bidirectional Long Short-Term Memory (Bi-LSTM) and Gated Recurrent Unit (GRU)) combined with data balancing techniques (random oversampling and Tomek links) to mitigate data imbalance issue. To establish the effectiveness of the proposed models, the experiments were conducted on four code smells datasets (God class, data Class, feature envy, and long method) extracted from 74 open-source systems. We compare and evaluate the performance of the models according to seven different performance measures accuracy, precision, recall, f-measure, Matthew’s correlation coefficient (MCC), the area under a receiver operating characteristic curve (AUC), the area under the precision–recall curve (AUCPR) and mean square error (MSE). After comparing the results obtained by the proposed models on the original and balanced data sets, we found out that the best accuracy of 98% was obtained for the Long method by using both models (Bi-LSTM and GRU) on the original datasets, the best accuracy of 100% was obtained for the long method by using both models (Bi-LSTM and GRU) on the balanced datasets (using random oversampling), and the best accuracy 99% was obtained for the long method by using Bi-LSTM model and 99% was obtained for the data class and Feature envy by using GRU model on the balanced datasets (using Tomek links). The results indicate that the use of data balancing techniques had a positive effect on the predictive accuracy of the models presented. The results show that the proposed models can detect the code smells more accurately and effectively.

Comparative Analysis of Deep Learning Models for Part of Speech Tagging in the Malay Language

Despite the widespread use of Malay, under-resourced languages like Malay face challenges in Natural Language Processing (NLP), particularly in Part-of-Speech (POS) tagging. The scarcity of annotated corpora poses a primary obstacle to POS tagging in Malay. This study aims to enhance the effectiveness and reliability of POS tagging models explicitly tailored for under-resourced languages within the field of NLP, focusing on Malay. Existing models, which rely on Conditional Random Fields and Hidden Markov Models, exhibit limitations, underscoring the need for more robust approaches. The research conducts a comparative analysis of various deep-learning models with different encoders for POS tagging in Malay sentences. The experimental analysis demonstrates that the Bidirectional Long Short-Term Memory (Bi-LSTM) model, leveraging a pre-trained Bidirectional Encoder Representations from Transformers (BERT) embedding model, achieves exceptional accuracy, precision, recall, and F1 scores in predicting tags. Notably, the BERT + Bi-LSTM model, boasting an accuracy of 98.82%, outperforms other models, showcasing superior performance across all evaluated metrics. Additionally, this combined model effectively handles known and unknown words, yielding highly accurate POS tagging results for Malay sentences. Doi: 10.28991/HIJ-2024-05-02-04 Full Text: PDF

Review on Sanskrit Sandhi Splitting using Deep Learning Techniques

A procedure called sandhi is used in Sanskrit to join short words (morphemes) to create compound words. A composite word are broken down into their component morphemes by a process known as sandhi splitting. This study focuses on several performance technologies and methodologies used to perform the above operation on Sanskrit sentences. Various approaches were identified for the problem from the literature survey. Initial approaches involved use of Finite State Transducers. Earlier the approaches introduced to increase accuracy include use of mathematical models and various optimality theories. Graph based approaches and parser based techniques were introduced later. With the advancement of deep learning techniques Recurrent Neural Networks, Long-Short Term Memory models and Double decoder models were adopted which involved training machine learning models through neural networks and classifier algorithms. Bidirectional LSTM models with attention mechanism, transformer based models and large language models like BERT were the most recent methodologies adopted and proved to be of higher accuracy and performance.

An Ensemble Machine Learning Technique for Bitcoin Price Prediction

This research proposes an ensemble approach for Bitcoin price prediction, leveraging historical price data and sentiment analysis. The proposed ensemble approach combines the model with Gated Recurrent Unit (GRU) and Bidirectional Long Short-Term Memory (BiLSTM) to further improve the accuracy in prediction by considering dynamics in the market. The model also addresses the problem of generalization and overfitting, adaption to the changing, dynamic nature of the market. Historical price data and sentiment scores from the preprocessing of the text are combined to the ensemble framework. These data are then fed into GRU and BiLSTM models for training, as the data contain not only complex temporal patterns but also sentiment-driven trends. The ensemble strategy could be beneficial for the strengths of the models and for improving the performances of the predictors. Most importantly, features are engineered in terms of technical indicators, lagged variables, and external factors impacting the price of Bitcoin. Sentiment analysis with the news and on social media complements insight into market sentiment, which adds value to the prediction power of the model.