Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

COVID-19 is a pandemic that has affected over 170 countries around the world. The number of infected and deceased patients has been increasing at an alarming rate in almost all the affected nations. Forecasting techniques can be inculcated thereby assisting in designing better strategies and in taking productive decisions. These techniques assess the situations of the past thereby enabling better predictions about the situation to occur in the future. These predictions might help to prepare against possible threats and consequences. Forecasting techniques play a very important role in yielding accurate predictions. This study categorizes forecasting techniques into two types, namely, stochastic theory mathematical models and data science/machine learning techniques. Data collected from various platforms also play a vital role in forecasting. In this study, two categories of datasets have been discussed, i.e., big data accessed from World Health Organization/National databases and data from a social media communication. Forecasting of a pandemic can be done based on various parameters such as the impact of environmental factors, incubation period, the impact of quarantine, age, gender and many more. These techniques and parameters used for forecasting are extensively studied in this work. However, forecasting techniques come with their own set of challenges (technical and generic). This study discusses these challenges and also provides a set of recommendations for the people who are currently fighting the global COVID-19 pandemic.

/pdf/forecasting-models-for-coronavirus-disease-covid-19-a-survey-40j1pyy537.pdf

Forecasting Models for Coronavirus Disease (COVID-19): A Survey of the State-of-the-Art

Stock market prediction has been identified as a very important practical problem in the economic field. However, the timely prediction of the market is generally regarded as one of the most challenging problems due to the stock market’s characteristics of noise and volatility. To address these challenges, we propose a deep learning-based stock market prediction model that considers investors’ emotional tendency. First, we propose to involve investors’ sentiment for stock prediction, which can effectively improve the model prediction accuracy. Second, the stock pricing sequence is a complex time sequence with different scales of fluctuations, making the accurate prediction very challenging. We propose to gradually decompose the complex sequence of stock price by adopting empirical modal decomposition (EMD), which yields better prediction accuracy. Third, we adopt LSTM due to its advantages of analyzing relationships among time-series data through its memory function. We further revised it by adopting attention mechanism to focus more on the more critical information. Experiment results show that the revised LSTM model can not only improve prediction accuracy, but also reduce time delay. It is confirmed that investors’ emotional tendency is effective to improve the predicted results; the introduction of EMD can improve the predictability of inventory sequences; and the attention mechanism can help LSTM to efficiently extract specific information and current mission objectives from the information ocean.

Stock closing price prediction based on sentiment analysis and LSTM

Stock-market prediction using machine-learning technique aims at developing effective and efficient models that can provide a better and higher rate of prediction accuracy. Numerous ensemble regressors and classifiers have been applied in stock market predictions, using different combination techniques. However, three precarious issues come in mind when constructing ensemble classifiers and regressors. The first concerns with the choice of base regressor or classifier technique adopted. The second concerns the combination techniques used to assemble multiple regressors or classifiers and the third concerns with the quantum of regressors or classifiers to be ensembled. Subsequently, the number of relevant studies scrutinising these previously mentioned concerns are limited. In this study, we performed an extensive comparative analysis of ensemble techniques such as boosting, bagging, blending and super learners (stacking). Using Decision Trees (DT), Support Vector Machine (SVM) and Neural Network (NN), we constructed twenty-five (25) different ensembled regressors and classifiers. We compared their execution times, accuracy, and error metrics over stock-data from Ghana Stock Exchange (GSE), Johannesburg Stock Exchange (JSE), Bombay Stock Exchange (BSE-SENSEX) and New York Stock Exchange (NYSE), from January 2012 to December 2018. The study outcome shows that stacking and blending ensemble techniques offer higher prediction accuracies (90–100%) and (85.7–100%) respectively, compared with that of bagging (53–97.78%) and boosting (52.7–96.32%). Furthermore, the root means square error (RMSE) recorded by stacking (0.0001–0.001) and blending (0.002–0.01) shows a better fit of ensemble classifiers and regressors based on these two techniques in market analyses compared with bagging (0.01–0.11) and boosting (0.01–0.443). Finally, the results undoubtedly suggest that an innovative study in the domain of stock market direction prediction ought to include ensemble techniques in their sets of algorithms.

/pdf/a-comprehensive-evaluation-of-ensemble-learning-for-stock-14pcf72a0y.pdf

A comprehensive evaluation of ensemble learning for stock-market prediction

This work presents and assesses the power of various volumetric, sentiment, and social network approaches to predict crucial decisions from online social media platforms. The views of individuals play a vital role in the discovery of some critical decisions. Social media has become a well-known platform for voicing the feelings of the general population around the globe for almost decades. Sentiment analysis or opinion mining is a method that is used to mine the general population’s views or feelings. In this respect, the forecasting of election results is an application of sentiment analysis aimed at predicting the outcomes of an ongoing election by gauging the mood of the public through social media. This survey paper outlines the evaluation of sentiment analysis techniques and tries to edify the contribution of the researchers to predict election results through social media content. This paper also gives a review of studies that tried to infer the political stance of online users using social media platforms such as Facebook and Twitter. Besides, this paper highlights the research challenges associated with predicting election results and open issues related to sentiment analysis. Further, this paper also suggests some future directions in respective election prediction using social media content.

The emergence of social media data and sentiment analysis in election prediction

This research work emphases on the prediction of future stock market index values based on historical data. The experimental evaluation is based on historical data of 10 years of two indices, namely, CNX Nifty and S&P Bombay Stock Exchange (BSE) Sensex from Indian stock markets. The predictions are made for 1–10, 15, 30, and 40 days in advance. This work proposes to combine the predictions/estimates of the ensemble of trees in a Random Forest using LSboost (i.e. LS-RF). The prediction performance of the proposed model is compared with that of well-known Support Vector Regression. Technical indicators are selected as inputs to each of the prediction models. The closing value of the stock price is the predicted variable. Results show that the proposed scheme outperforms Support Vector Regression and can be applied successfully for building predictive models for stock prices prediction.

Combining of random forest estimates using LSboost for stock market index prediction

Swine flu, a respiratory disease caused by influenza viruses also NIHI virus, infects pig’s respiratory tract which results in a barking cough, decrease in appetite, secretions of nasal and endless behavior. These viruses may mutate so that they are easily transmissible among humans. To deal with these viruses, in this manuscript we have used various time series forecasting models to predict future cases of swine flu in India so that Government could take necessary steps to prevent the spread of this disease. Swine Flu data is collected from integrated diseases surveillance programme, Government of India. The data is of year 2010 to 2017; which provides information on yearly cases of Swine flu in different states of India. The forecasting models applied are Box-Cox transformation, Exponential Smoothing, Seasonal Naive and Neural Network. The result of the applied models are compared on the basis of errors such as Mean Error, Mean Absolute Error, Root Mean Square Error, Mean Absolute Scaled Error and Auto Correlation Function. Analyzing the final result we realized that the Neural Network forecasting model gives the best result among all others with the accuracy of 98.4%.

Statistical Models for Predicting Swine F1u Incidences in India

This study aims to explore the impact of heterogeneity on a hybrid algorithm called Multi Adaptive Filter Algorithm by constructing series of experiments. Here, the simulations were made between ‘Total Energy Spent’ and ‘Number of Sources’ considering temporal correlation. The results were drawn from the trace information generated using ‘Monte Carlo’ simulation methods. After keen analysis, the results show that different levels of heterogeneity are best suited for correlated event detections. Moreover, based on the conclusions drawn, it can be safely inferred that n-level heterogeneity reduces the total energy spent close to 60%. Further, cost analysis recommends that adding progressive nodes preserves the cost factor in the bracket of 230–280$/Joule. The novel approach can immensely help the future solution providers to overcome the battery limitations of wireless sensor networks. This study provides insights into designing heterogeneous Wireless Sensor Networks and aims at providing the cost-benefit analysis that can be used in selecting the critical parameters of the network.

Cost analysis of hybrid adaptive routing protocol for heterogeneous wireless sensor network

In Recent times, Chikungunya is considered as one of the most severe disease in India. It is caused by mosquitoes bite (CHIKV). But till now around the globe, scientists are unable to find the exact cure of this disease. Hence as a precautionary measure, there is an imperative need to predict the future possibilities of Chikungunya cases. Therefore, in this manuscript, machine learning based forecasting models are used for prediction of chikungunya cases in India for year 2018-2024. Analysis is conducted on the data of past years (2007-2017) Chikungunya cases in India. Box Cox, Mean Forecast, Seasonal Naive, and Neural Network are techniques are used for analysis and forecasting. The surpassing model is adopted based on the accuracy factor. Accuracy of the models are compared with respect to Root Mean Square Error and Auto Correlation Function. Result analysis reveal that the neural network model produces least error and hence is the best prediction model for our dataset in terms of accuracy.

Statistical Models for Predicting Chikungunya Incidences in India

Sentiment analysis is all about finding polarity (positivity and negativity) of a given text which can be a single word, a single sentence, a single document and maybe a whole book, novel or magazine. In this paper, sentiment analysis of the harry potter series is done using “tidytext” package in R. The “tidytext” package includes three lexicons and this paper discusses sentiment analysis results encountered by them on same data. This study has analyzed results of all three lexicons for the harry potter series and found that overall sentiments of the harry potter series are negative on the basis of positive and negative words present in the series.

Nonita Sharma

Papers

Combining of random forest estimates using LSboost for stock market index prediction

Statistical Models for Predicting Swine F1u Incidences in India

Cost analysis of hybrid adaptive routing protocol for heterogeneous wireless sensor network

Statistical Models for Predicting Chikungunya Incidences in India

Sentiment Analysis using tidytext package in R