Like Long Short-Term Memory Research Articles

Air Quality Prediction using Ensemble Voting based Deep Learning with Mud Ring Algorithm for Intelligent Transportation Systems

<p>In recent times, advanced technologies in transportation are developing like connected and automated vehicles and shared mobility services. A rapidly increasing number of vehicles in intelligent transportation system (ITS) and smart cities causes pollution and degrade the quality of air. Owing to the incredible effect of air quality on individual lives, it is indispensable to design a system by which air pollutants (PM2.5, NOx, COx, SOx) are predicted. But predicting air quality and its pollutants was complex since air quality relies on various elements like power plants, weather, and vehicular emissions. Deep learning (DL) and Machine learning (ML) techniques are leveraged for developing an air quality predictive method. This study develops an Air Quality Prediction utilizing Ensemble Voting based Deep Learning with Mud Ring Algorithm (AQP-EDLMRA) technique. The presented AQP-EDLMRA technique follows the ensemble voting model, which exploits three DL classification methods like long short-term memory (LSTM), deep belief network (DBN), and stacked autoencoder (SAE). Then, the new data can be classified by the weighted vote of their prediction outcomes. To adjust the hyperparameter values of the DL methods, the MRA was exploited, showing the novelty of the work. The simulation values of the AQP-EDLMRA approach are tested using a series of air quality data and the comprehensive comparative results demonstrated that the AQP-EDLMRA technique has reached improved forecasting performance.</p>

Particle swarm optimization based LSTM networks for water level forecasting: A case study on Bangladesh river network

Floods are one of the most catastrophic natural disasters. Water level forecasting is an essential method of avoiding floods and disaster preparedness. In recent years, models for predicting water levels have been developed using artificial intelligence techniques like the artificial neural network (ANN). It has been demonstrated that more advanced and sequenced-based deep learning techniques, like long short-term memory (LSTM) networks, are superior at forecasting hydrological data. However, historically, most LSTM hyperparameters were based on experience, which typically did not produce the best outcomes. The Particle Swarm Optimization (PSO) method was utilized to adjust the LSTM hyperparameter to increase the capacity to learn data sequence characteristics. Utilizing water level observation data from stations along Bangladesh's Brahmaputra, Ganges, and Meghna rivers, the model was utilized to estimate flood dynamics. The Nash Sutcliffe efficiency (NSE) coefficient, root mean square error (RMSE), and MAE were used to assess the model's performance, where PSO-LSTM model outperforms the ANN, PSO-ANN, and LSTM models in predicting water levels in all stations. The PSO-LSTM model provides improved prediction accuracy and stability and improves water level forecasting accuracy at varying lead times. The findings may aid in sustainable flood risk mitigation in the study region in the future.

LSTM projected layer neural network-based signal estimation and channel state estimator for OFDM wireless communication systems

<abstract><p>Advanced wireless communication technologies, such as 5G, are faced with significant challenges in accurately estimating the transmitted signal and characterizing the channel. One of the major obstacles is the interference caused by the delay spread, which results from receiving multiple signal copies through different paths. To mitigate this issue, the orthogonal frequency division modulation (OFDM) technique is often employed. Efficient signal detection and optimal channel estimation are crucial for enhancing the performance of multi-carrier wireless communication systems. To this end, this paper proposes a Long Short Term Memory-Projected Layer (LSTM-PL) deep neural network(DNN) based channel estimator to detect received OFDM signal. The results show that the LSTM-PL algorithm outperforms traditional methods such as Least Squares(LS), Minimum Mean Square Error (MMSE) and other LSTM deep learning channel estimation methods like Long Short Term Memory(LSTM)-DNN and Bidirectional-LSTM(Bi-LSTM)-DNN, as evidenced by Symbol-Error Rate (SER) outcomes.</p></abstract>

Application of LiDAR Data for Deep Learning Based Near Crash Prediction at Signalized Intersection

Near crash events are often regarded as an excellent surrogate measure for traffic safety research because they include abrupt changes in vehicle kinematics that can lead to deadly accident scenarios. In this paper, we introduced machine learning and deep learning algorithms for predicting near crash events using LiDAR data at a signalized intersection. To predict a near crash occurrence, we used essential vehicle kinematic variables such as lateral and longitudinal velocity, yaw, tracking status of LiDAR, etc. A deep learning hybrid model Convolutional Gated Recurrent Neural Network (CNN + GRU) was introduced, and comparative performances were evaluated with multiple machine learning classification models such as Logistic Regression, K Nearest Neighbor, Decision Tree, Random Forest, Adaptive Boost, and deep learning models like Long Short-Term Memory (LSTM). As vehicle kinematics changes occur after sudden brake, we considered average deceleration and kinematic energy drop as thresholds to identify near crashes after vehicle braking time . We looked at the next 3 seconds of this braking time as our prediction horizon. All models work best in the next 1-second prediction horizon to braking time. The results also reveal that our hybrid model gathers the greatest near crash information while working flawlessly. In comparison to existing models for near crash prediction, our hybrid Convolutional Gated Recurrent Neural Network model has 100% recall, 100% precision, and 100% F1-score: accurately capturing all near crashes. This prediction performance outperforms previous baseline models in forecasting near crash events and provides opportunities for improving traffic safety via Intelligent Transportation Systems (ITS).

Semi-supervised LSTM with historical feature fusion attention for temporal sequence dynamic modeling in industrial processes

In modern industrial processes, the data-driven soft sensor technology has been widely used for the prediction of key quality variables. Due to the important of dynamics and nonlinearity in industrial process data, deep learning models like long short-term memory (LSTM) network are well suited for temporal sequence dynamic modeling due to their excellent long-term memory function and feature extraction capability. Furthermore, industrial processes generate a large amount of process data with irregular sampling frequencies. However, traditional LSTM cannot fully utilize the process data with irregular sampling frequency and the guidance value of historical data samples for feature learning. To address these issues, a novel semi-supervised LSTM with history feature fusion attention (HFFA-SSLSTM) model is proposed in this paper. First, the semi-supervised learning strategy is implemented in LSTM to fully utilize the unlabeled data and mine the temporal sequence features of labeled samples and unlabeled samples with irregular sampling frequencies. Then, a novel historical feature fusion attention (HFFA) mechanism is developed, which utilizes historical hidden features to learn attention scores for obtaining weighted historical information-related features. Finally, the extracted features are combined to form the soft sensor model to perform time series prediction tasks for key quality variables in industrial processes. The experimental results on the actual industrial hydrocracking data set demonstrate the effectiveness of the proposed HFFA-SSLSTM model and its possibility of applicating in real industrial processes.

Air Quality Index prediction using an effective hybrid deep learning model

Environmentalism has become an intrinsic part of everyday life. One of the greatest challenge to the environment's long-term existence is the air pollution. Delhi, the capital of India, has experienced decreasing of air quality for several years. The poor air quality has a significant impact on the lives of individuals. Air Quality Index (AQI) prediction can help to its beneficiaries in taking safeguards about their health before moving to any polluted area. In this study, a variety of data forecasting approaches is evaluated to predict the AQI value for Particulate Matter (PM2.5) μm at a particular area of Delhi and several error-prone strategies such as R-Squared (R2), Mean Absolute Error (MAE), and Root Mean Square Error (RMSE) methods are catalogued. In the proposed approach two deep learning models like Long-Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) are combined to predict the AQI of the environment. Several stand alone machine learning (ML) and deep learning (DL) models such as LSTM, Linear-Regression (LR), GRU, K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) are also trained on the same dataset to compare their performances with the proposed hybrid (LSTM-GRU) model and it is found that the proposed hybrid model shows supremacy in the performance with the MAE value 36.11 and R2 value 0.84.

Differentiable, Learnable, Regionalized Process‐Based Models With Multiphysical Outputs can Approach State‐Of‐The‐Art Hydrologic Prediction Accuracy

AbstractPredictions of hydrologic variables across the entire water cycle have significant value for water resources management as well as downstream applications such as ecosystem and water quality modeling. Recently, purely data‐driven deep learning models like long short‐term memory (LSTM) showed seemingly insurmountable performance in modeling rainfall runoff and other geoscientific variables, yet they cannot predict untrained physical variables and remain challenging to interpret. Here, we show that differentiable, learnable, process‐based models (called δ models here) can approach the performance level of LSTM for the intensively observed variable (streamflow) with regionalized parameterization. We use a simple hydrologic model HBV as the backbone and use embedded neural networks, which can only be trained in a differentiable programming framework, to parameterize, enhance, or replace the process‐based model's modules. Without using an ensemble or post‐processor, δ models can obtain a median Nash‐Sutcliffe efficiency of 0.732 for 671 basins across the USA for the Daymet forcing data set, compared to 0.748 from a state‐of‐the‐art LSTM model with the same setup. For another forcing data set, the difference is even smaller: 0.715 versus 0.722. Meanwhile, the resulting learnable process‐based models can output a full set of untrained variables, for example, soil and groundwater storage, snowpack, evapotranspiration, and baseflow, and can later be constrained by their observations. Both simulated evapotranspiration and fraction of discharge from baseflow agreed decently with alternative estimates. The general framework can work with models with various process complexity and opens up the path for learning physics from big data.

PREDICTING SATELLITE TELEMETRY HEALTH PARAMETERS USING MACHINE LEARNING

In this paper, a methodology is used for predicting time series data which can be used for predicting spacecraft health parameters during non-visible period using the last received health parameters from the spacecraft (simulated parameters). Machine Learning based methodology to monitor telemetry parameters for early detection of anomalies and predicting trends is used. Explored different models and comparing them to see which one gives the best accuracy. We have proposed Machine learning using supervised learning algorithms like Long Short-Term Memory (LSTM) approach as it is well suited for solving this type of regression problem. LSTM is a special case of Recurrent Neural Networks (RNN) which is capable of learning long-term dependencies.

Deep Learning Paradigm for Cardiovascular Disease/Stroke Risk Stratification in Parkinson's Disease Affected by COVID-19: A Narrative Review.

Background and Motivation: Parkinson’s disease (PD) is one of the most serious, non-curable, and expensive to treat. Recently, machine learning (ML) has shown to be able to predict cardiovascular/stroke risk in PD patients. The presence of COVID-19 causes the ML systems to become severely non-linear and poses challenges in cardiovascular/stroke risk stratification. Further, due to comorbidity, sample size constraints, and poor scientific and clinical validation techniques, there have been no well-explained ML paradigms. Deep neural networks are powerful learning machines that generalize non-linear conditions. This study presents a novel investigation of deep learning (DL) solutions for CVD/stroke risk prediction in PD patients affected by the COVID-19 framework. Method: The PRISMA search strategy was used for the selection of 292 studies closely associated with the effect of PD on CVD risk in the COVID-19 framework. We study the hypothesis that PD in the presence of COVID-19 can cause more harm to the heart and brain than in non-COVID-19 conditions. COVID-19 lung damage severity can be used as a covariate during DL training model designs. We, therefore, propose a DL model for the estimation of, (i) COVID-19 lesions in computed tomography (CT) scans and (ii) combining the covariates of PD, COVID-19 lesions, office and laboratory arterial atherosclerotic image-based biomarkers, and medicine usage for the PD patients for the design of DL point-based models for CVD/stroke risk stratification. Results: We validated the feasibility of CVD/stroke risk stratification in PD patients in the presence of a COVID-19 environment and this was also verified. DL architectures like long short-term memory (LSTM), and recurrent neural network (RNN) were studied for CVD/stroke risk stratification showing powerful designs. Lastly, we examined the artificial intelligence bias and provided recommendations for early detection of CVD/stroke in PD patients in the presence of COVID-19. Conclusion: The DL is a very powerful tool for predicting CVD/stroke risk in PD patients affected by COVID-19.

Applying the Spatial Transmission Network to the Forecast of Infectious Diseases Across Multiple Regions.

ObjectiveTimely and accurate forecast of infectious diseases is essential for achieving precise prevention and control. A good forecasting method of infectious diseases should have the advantages of interpretability, feasibility, and forecasting performance. Since previous research had illustrated that the spatial transmission network (STN) showed good interpretability and feasibility, this study further explored its forecasting performance for infectious diseases across multiple regions. Meanwhile, this study also showed whether the STN could overcome the challenges of model rationality and practical needs.MethodsThe construction of the STN framework involved three major steps: the spatial kluster analysis by tree edge removal (SKATER) algorithm, structure learning by dynamic Bayesian network (DBN), and parameter learning by the vector autoregressive moving average (VARMA) model. Then, we evaluated the forecasting performance of STN by comparing its accuracy with that of the mechanism models like susceptible-exposed-infectious-recovered-susceptible (SEIRS) and machine-learning algorithm like long-short-term memory (LSTM). At the same time, we assessed the robustness of forecasting performance of STN in high and low incidence seasons. The influenza-like illness (ILI) data in the Sichuan Province of China from 2010 to 2017 were used as an example for illustration.ResultsThe STN model revealed that ILI was likely to spread among multiple cities in Sichuan during the study period. During the whole study period, the forecasting accuracy of the STN (mean absolute percentage error [MAPE] = 31.134) was significantly better than that of the LSTM (MAPE = 41.657) and the SEIRS (MAPE = 62.039). In addition, the forecasting performance of STN was also superior to those of the other two methods in either the high incidence season (MAPE = 24.742) or the low incidence season (MAPE = 26.209), and the superiority was more obvious in the high incidence season.ConclusionThis study applied the STN to the forecast of infectious diseases across multiple regions. The results illustrated that the STN not only had good accuracy in forecasting performance but also indicated the spreading directions of infectious diseases among multiple regions to a certain extent. Therefore, the STN is a promising candidate to improve the surveillance work.

Study and Analysis of Stock Market Prediction Techniques

Stock marketplace is a complicated and demanding system in which people make more money or lose their entire savings. The stock market prediction having high accuracy yields more profit for stock investors. Stock market data is generated in a very large amount and it varies quickly every second. The decision making in stock marketplace is a very challenging and strenuous task of financial stock market. The development of efficient models for prediction decisions is very difficult because of the convolution of stock market financial data and should have high accuracy. This study attempts to compare existing models for the stock market. Various Machine learning methods like Long Short Term Memory (LSTM), Convolution Neural Networks (CNN) and Convolution Neural Networks – Long Term Short Memory (CNN-LSTM) have been used for the comparison. The models are estimated using conventional strategic measure: MAE (Mean Absolute Error). The measured low values indicates that the models are effective in predicting stock prices.

Attention-Based Multi-Learning Approach for Speech Emotion Recognition With Dilated Convolution

The success of deep learning in speech emotion recognition has led to its application in resource-constrained devices. It has been applied in human-to-machine interaction applications like social living assistance, authentication, health monitoring and alertness systems. In order to ensure a good user experience, robust, accurate and computationally efficient deep learning models are necessary. Recurrent neural networks (RNN) like long short-term memory (LSTM), gated recurrent units (GRU) and their variants that operate sequentially are often used to learn time series sequences of the signal, analyze long-term dependencies and the contexts of the utterances in the speech signal. However, due to their sequential operation, they encounter problems in convergence and sluggish training that uses a lot of memory resources and encounters the vanishing gradient problem. In addition, they do not consider spatial cues that may exist in the speech signal. Therefore, we propose an attention-based multi-learning model (ABMD) that uses residual dilated causal convolution (RDCC) blocks and dilated convolution (DC) layers with multi-head attention. The proposed ABMD model achieves comparable performance while taking global contextualized long-term dependencies between features in a parallel manner using a large receptive field with less increase in the number of parameters compared to the number of layers and considers spatial cues among the speech features. Spectral and voice quality features extracted from the raw speech signals are used as inputs. The proposed ABMD model obtained a recognition accuracy and F1 score of 93.75% and 92.50% on the SAVEE datasets, 85.89% and 85.34% on the RAVDESS datasets and 95.93% and 95.83% on the EMODB datasets. The model’s robustness in terms of the confusion ratio of the individual discrete emotions especially happiness which is often confused with emotions that belong to the same dimensional plane with it also improved when validated on the same datasets.

Forecasting Electricity Load With Hybrid Scalable Model Based on Stacked Non Linear Residual Approach

Power has totally different attributes than other material commodities as electrical energy stockpiling is a costly phenomenon. Since it should be generated when demanded, it is necessary to forecast its demand accurately and efficiently. As electrical load data is represented through time series pattern having linear and non-linear characteristics, it needs a model that may handle this behavior well in advance. This paper presents a scalable and hybrid approach for forecasting the power load based on Vector Auto Regression (VAR) and hybrid deep learning techniques like Long Short Term Memory (LSTM) and Convolutional Neural Network (CNN). CNN and LSTM models are well known for handling time series data. The VAR model separates the linear pattern in time series data, and CNN-LSTM is utilized to model non-linear patterns in data. CNN-LSTM works as CNN can extract complex features from electricity data, and LSTM can model temporal information in data. This approach can derive temporal and spatial features of electricity data. The experiment established that the proposed VAR-CNN-LSTM(VACL) hybrid approach forecasts better than more recent deep learning methods like Multilayer Perceptron (MLP), CNN, LSTM, MV-KWNN, MV-ANN, Hybrid CNN-LSTM and statistical techniques like VAR, and Auto Regressive Integrated Moving Average (ARIMAX). Performance metrics such as Mean Square Error, Root Mean Square Error, and Mean Absolute Error have been used to evaluate the performance of the discussed approaches. Finally, the efficacy of the proposed model is established through comparative studies with state-of-the-art models on Household Power Consumption Dataset (UCI machine learning repository) and Ontario Electricity Demand dataset (Canada).

Hybrid and Ensemble Methods of Two Days Ahead Forecasts of Electric Energy Production in a Small Wind Turbine

The ability to forecast electricity generation for a small wind turbine is important both on a larger scale where there are many such turbines (because it creates problems for networks managed by distribution system operators) and for prosumers to allow current energy consumption planning. It is also important for owners of small energy systems in order to optimize the use of various energy sources and facilitate energy storage. The research presented here addresses an original, rarely predicted 48 h forecasting horizon for small wind turbines. This topic has been rather underrepresented in research, especially in comparison with forecasts for large wind farms. Wind speed forecasts with a 48 h horizon are also rarely used as input data. We have analyzed the available data to identify potentially useful explanatory variables for forecasting models. Eight sets with increasing data amounts were created to analyze the influence of the types and amounts of data on forecast quality. Hybrid, ensemble and single methods are used for predictions, including machine learning (ML) solutions like long short-term memory (LSTM), multi-layer perceptron (MLP), support vector regression (SVR) and K-nearest neighbours regression (KNNR). Original hybrid methods, developed for research of specific implementations and ensemble methods based on hybrid methods’ decreased errors of energy generation forecasts for small wind turbines in comparison with single methods. The “artificial neural network (ANN) type MLP as an integrator of ensemble based on hybrid methods” ensemble forecasting method incorporates an original combination of predictors. Predictions by this method have the lowest mean absolute error (MAE). In addition, this paper presents an original ensemble forecasting method, called “averaging ensemble based on hybrid methods without extreme forecasts”. Predictions by this method have the lowest root mean square error (RMSE) error among all tested methods. LSTM, a deep neural network, is the best single method, MLP is the second best one, while SVR, KNNR and, especially, linear regression (LR) perform less well. We prove that lagged values of forecasted time series slightly increase the accuracy of predictions. The same applies to seasonal and daily variability markers. Our studies have also demonstrated that using the full set of available input data and the best proposed hybrid and ensemble methods yield the lowest error. The proposed hybrid and ensemble methods are also applicable to other short-time generation forecasting in renewable energy sources (RES), e.g., in photovoltaic (PV) systems or hydropower.

Sampling-Interval-Aware LSTM for Industrial Process Soft Sensing of Dynamic Time Sequences With Irregular Sampling Measurements

In modern industrial processes, dynamics and nonlinearities are two main difficulties for soft sensing of key quality variables. Thus, nonlinear dynamic models like long short-term memory (LSTM) network have been applied for data sequence modeling due to its powerful representation ability. Nevertheless, most dynamic methods cannot deal with data series with irregular sampling intervals, which is a common phenomenon in many industrial plants. To handle this problem, a novel sampling-interval-aware LSTM (SIA-LSTM) is proposed in this paper, which takes the sampling intervals between sequential samples into consideration to adjust the influence of the previous sample on the current one. To this end, two non-increasing functions of the sampling interval are designed to weight different sampling intervals in the dynamic data sequence. Then, each sampling-interval weight is multiplied to the corresponding previous hidden state to adjust its impact. Finally, the adjusted hidden state is used as an adaptive input for the three control gates in each LSTM unit to obtain the current hidden state. The proposed SIA-LSTM is applied to an actual hydrocracking process for soft sensor of the C5 content in the light naphtha and the final boiling point of the heavy naphtha.

Detecting Hate Speech using Deep Learning Techniques

Social networking sites saw a steep rise in terms of number of users in last few years. As a result of this, the interaction among the users also increased considerably. Along with these posting racial comments based on cast, race, gender, religion, etc. also increased. This propagation of negative messages is collectively known as hate speeches. Often these posts containing negative comments in social networking sites create law and order situations in the society, leading to loss of human life and properties. Detecting hate speech is one of the major challenges faced in recent time. In recent past, there have been a considerable amount of research going on the field of detection of hate speech in the social networking sites. Researchers in the fields of Natural Language Processing and Machine Learning have done considerable amount research in in this area. This paper uses a simple up sampling method to make the data balanced and implements deep learning models like Long Short Term Memory (LSTM) and Bi-directional Long Short Term Memory (Bi-LSTM) for improved accuracy in detecting hate speech in social networking sites. LSTM was found to have better accuracy that Bi-LSTM for the data set considered. LSTM also had better values for precision and F1 score. Bi-LSTM only for higher values for recall.

A Deep Learning-based Approach for Stock Price Prediction

Predicting stock prices accurately is difficult owing to the numerous elements that might influence stock prices, such as economic indicators, news events, and market patterns. Conventional approaches to stock price prediction frequently depend on statistical models that may be incapable of capturing the complicated interactions between these variables. Deep learning approaches like Long Short-Term Memory (LSTM) networks have showed promise in understanding these complicated linkages and effectively forecasting stock values in recent years. We conduct a literature review on deep learning-based techniques for stock price prediction in this study. We give an overview of deep learning techniques and how they may be used to forecast stock prices. We investigate publically available datasets for training and evaluating deep learning models for stock price prediction. We also show a case study in which we anticipate the stock values of many businesses listed on the National Stock Exchange of India using an LSTM-based model. Our findings show that the LSTM-based model outperforms traditional approaches to stock price prediction, showing that deep learning techniques have the potential to be a valuable tool for investors looking to make educated stock investing decisions.

Sea Level Anomaly and Earthquake Predictions: Endangered Countries Prognostications

Natural calamities are often unforeseen and cause massive destruction. It is extremely difficult to predict natural disasters. Existing machine learning techniques are not reliable enough to find the affected countries due to earthquakes and rising sea levels. The aim of this paper is to use predictive analysis to find the countries that will be affected by earthquakes and rising sea levels. Also, the purpose is to see how machine learning techniques perform in terms of sudden calamities like earthquakes or slow calamities like rising sea level. The results was deduced by data analysis, and deep learning techniques like Long-Short Term Memory (LSTM). It was found out that using the approached method in this paper can accurately identify the countries that are going to be affected and predict both earthquake and sea level anomalies accurately. For earthquake, the model was able to capture the happening of earthquake events into a certain quarter of the year with the Root Mean Square Error (RMSE) of 0.504. And for sea level rise, the RMSE was 0.064. It was concluded that Deep learning techniques (e.g.-LSTM) work well with slow changes like sea level anomaly rather than earthquakes. The techniques used in this paper can be upgraded further in the future to find and help more endangered countries to be prepared better against these sudden calamities.

Feature engineering combined with 1-D convolutional neural network for improved mortality prediction

Abstract Objectives The appropriate care for patients admitted in Intensive care units (ICUs) is becoming increasingly prominent, thus recognizing the use of machine learning models. The real-time prediction of mortality of patients admitted in ICU has the potential for providing the physician with the interpretable results. With the growing crisis including soaring cost, unsafe care, misdirected care, fragmented care, chronic diseases and evolution of epidemic diseases in the domain of healthcare demands the application of automated and real-time data processing for assuring the improved quality of life. The intensive care units (ICUs) are responsible for generating a wealth of useful data in the form of Electronic Health Record (EHR). This data allows for the development of a prediction tool with perfect knowledge backing. Method We aimed to build the mortality prediction model on 2012 Physionet Challenge mortality prediction database of 4,000 patients admitted in ICU. The challenges in the dataset, such as high dimensionality, imbalanced distribution and missing values, were tackled with analytical methods and tools via feature engineering and new variable construction. The objective of the research is to utilize the relations among the clinical variables and construct new variables which would establish the effectiveness of 1-Dimensional Convolutional Neural Network (1-D CNN) with constructed features. Results Its performance with the traditional machine learning algorithms like XGBoost classifier, Light Gradient Boosting Machine (LGBM) classifier, Support Vector Machine (SVM), Decision Tree (DT), K-Neighbours Classifier (K-NN), and Random Forest Classifier (RF) and recurrent models like Long Short-Term Memory (LSTM) and LSTM-attention is compared for Area Under Curve (AUC). The investigation reveals the best AUC of 0.848 using 1-D CNN model. Conclusion The relationship between the various features were recognized. Also, constructed new features using existing ones. Multiple models were tested and compared on different metrics.

Approaching Stock Market Trading with Echo State Networks

The US stock market is an integral part of modern society. Nearly 55% of Americans own corporate shares in the US stock market (What Percentage of Americans Own Stock?, 2019), and as of June 30th, 2020, the total value of the US stock market was over 35 trillion USD (Total Market Value of U.S. Stock Market, 2020). The stock market is also extremely volatile, and many people have gone bankrupt from poor investments. To minimize the risk and capitalize off the massive amounts of data on corporations and share prices present in the world, algorithmic trading began to rise. Trading algorithms have the potential for huge returns, and while many algorithms employ strategies like Long-Short Equity, very few attempt to use machine learning due to the unpredictable nature of the stock market. Many time series prediction models like autoregressive integrated moving average (ARIMA), and even neural networks like long short term memory (LSTMs) often fail when predicting stock market data, because unlike other time series data, the stock market is almost never univariate, or follows seasonal trends. However, where other models come short, echo state networks (ESNs) excel, due to their reservoir like computing model, which allows them to perform better on messy, non traditional time series data. Using a combination of ESNs to predict prices, and clustering we created an algorithm model that can predict trends with over 95% confidence, but had mixed results accurately predicting returns.