Bitcoin Price Prediction

Abstract

In this paper, we have proposed to predict the value of Bitcoin accurately considering the various parameters that affect the value of Bitcoin. Collecting information from various references and using papers in real time, I discovered the advantages and disadvantages of predicting the price of Bitcoin. Each paper has its own set of ways to predict the price of Bitcoin. Many papers have an accurate value but some do not, but time complexity is high in those predictions, so to reduce the complexity of the time here in this paper we use an algorithm linked to artificial intelligence called LASSO (at least one opt-out operator. Some papers have used different algorithms such as SVM (support vector machine), GLM, CNN (Convolutional Neural Networks), and RNN (Recurrent neural networks) which do not have good time management, but LASSO acquisition of results on a larger website is faster and faster so for this purpose, we find comparisons between other algorithms and the LASSO algorithm, this test paper helps future researchers to make an impact on their papers. The process takes place in the first paper moment of research, we aim to understand and discover everyday trends in the Bitcoin market while gaining insight into the relevant features surrounding Bitcoin price. Our data set contains various features related to the value of Bitcoin and the payment network throughout the years, which are recorded daily. We previously processed data, using other data mining techniques to reduce data noise. Then for the second minute of our study, we use available information, and we will predict the daily price change signal with the highest accuracy.

Introduction

I. INTRODUCTION

Bitcoin is a cryptographic currency used worldwide as an advanced installment or a guess objective. Bitcoin is categorized for example it is not owned by anyone. Bitcoin exchanges exist simply as they are not connected to any nation. Guessing it has to happen through different commercial institutions known as "bitcoin trades". This enables individuals to know how to sell/buy Bitcoins using different currencies. The biggest Bitcoin trader is Mt Gox. Bitcoins have been put in place as an advanced wallet that resembles a virtual financial balance. The maximum value of this exchange, timestamp details is placed in a place called the Blockchain. Each record in the blockchain is known as a square. Each square has a past tense information square. Blockchain information he shook. During the exchange the client's name is incorrectly uncovered, however, their bag ID was made open. The value of Bitcoin fluctuates as a stock even though it is the internal unexpected way. There are various statistics used for information on the exchange rate financial information. The parameters influencing Bitcoin are unusual. In this way, it is important to wait for Bitcoin estimates before the right business choice can be made. The cost of Bitcoin does not depend on the business times or government interventions are not at all the same as a securities exchange.

II. LITERATURE REVIEW

[1] This paper is about the program as well as software-based learning transfer design engineering. As an important occupation, you receive low-level art and emphasis in the real world, open jobs. These activities develop the scope and the size of the skills, not the ones associated with the content's special knowledge or skills, yet more than that world skills. [2] The goal of this clever student business is to show you how a customized machine model can expect the cost of cryptographic money on the opportunity to provide a complete measure of information as well as computer influence. Displays a chart with expected qualities. The new most popular composition of the kind of machine system that can help humanity by foresight the coming events. We have finally come to a point where predictions can be accurate and generated by reliance on factual information. Besides, with the rise of advanced crypto time, many heads turned to the business computer market. [3] This can be practiced by utilizing a progression of AI strategies and philosophies. The main aim of this paper is to find the actual Bitcoin price in US dollars that can be predicted. The Bitcoin Price should be found in the price index of the dataset. The problem will be solved by achieving a level of success through the high implementation of a Bayesian regression. To optimize recurrent neural network (RNN) and a Long Short Term Memory (LSTM) network. The LST Achieves the highest classification accuracy of 52% and an RMSE of 8%.

The popular ARIMA model for time series forecasting is implemented as a comparison to the deep learning models. As expected, the non-linear deep learning methods outperform the ARIMA forecast which performs very low. So, Finally, both learning models have resulted from the outcomes being a very low level of accuracy. This section gives an overview of prediction architecture and a survey on Bitcoin Price prediction by using the machine learning algorithm techniques highlighted accordingly to survey papers of price predictions base papers. AI is seen as computer use to consider when accessible data is used as statistics to process or help manage facts and information. While AI combines machine ideas, it needs human guidance. AI covers this important level of guessing to find a framework that works well yet has hidden times of knowledge. AI is generally new internal control Software development that provides a variety of informal assessment methods.

III. PREDICTION TECHNIQUES

A. Linear Regression Model

Linear regression is a linear approach to modeling the relationship between a dependent variable and independent variables. The case of a linear variable is called simple linear regression. This paper uses the regression of the line model of the relationship between dependent and one or more independent variants.

B. K-Nearest Neighbor

K-means creates k groups from a set of objects so that the members of a group are more similar and based on this data are clustered as normal, stressed, or highly stressed. We can compute the distance between two dependent and independent variables using some distance function d(x,y), where x,y are scenarios composed Number of features, such that x={x1,…,xN}, y={y1,…,yN}. Break the third major source of information in all the unthinkable background times 180s, 360s, and 720s. Enter the meaning of k group to repatriate more than 100 communities each temporary size, and later use the Entropy test to limit this up to 20 best / most flexible and best crowds. Use the second cost plan to get it comparing several highlights obtained using the strategy of going back to Bayesian. Going back fills up as a pursuit. check three vectors for the past cost of different period times (180s, 360s, and 720s). Each time, find comparisons between these vectors and use 20 best k means to design with its acquired value hop, so find the possible value change dp. Count in each section using different developmental work.

C. Naïve Bayes

Naïve Bayes techniques are a great deal of coordinated learning on applying Bayes' speculation with the "honest" supposition of opportunity between each pair of features. Their look has been greatly improved considering the innocent Bayes categories are very active well in some real situations. They need a limited rate of information preparation for critical assessment boundaries. Honest Bayes understudies and classifiers can be unbelievably speedy and appear differently about progressively present-day systems. The decoupling of the class-prohibitive component dispersals suggests that each movement can be uninhibitedly evaluated as a one-dimensional scattering. This along these lines decreases issues originating from the scourge of dimensionality.

Random Forests

Random Forests get the outfit learning framework where distinctive weak understudies are merged to make a strong understudy. It is a meta-estimator that fits various decision tree classifiers on various sub-primer of the enlightening assortment and uses averaging to improve the farsighted accuracy and authority overfitting.

Build three-time arrangement informational indexes for 30, 60, and 120 minutes (180, 360, 720 information focuses individually) going before the present information point at all focuses in time separately.

Run GLM/Random Forest on each of the two-time series data sets separately.

We get two separate linear models: M1, and M2 corresponding to each of the data sets. From M1, we can predict the price change at t, denoted ?P1. Similarly, we have ?P2 for M2.

IV. PROPOSED METHODOLOGY

A. Algorithm

Least Absolute shrinkage selection

Operator (LASSO): In estimations and AI, rope (least absolute shrinkage selection operator or LASSO) is the faith assessment framework that performs both variable choice and regularization to refresh the check exactness and interpretability of the legitimate model it produces.

Diverse tie assortments have been made to fix certain constraints of the fundamental strategy and to make the system dynamically huge for unequivocal issues. In every practical sense these emphasis on as for or using various sorts of conditions among the covariates.

B. Decision Tree

This is one of my preferred calculations and I use it at times. It is a kind of directed learning calculation that is for the most part utilized for order issues. Shockingly, it works for both clear-cut and consistent ward factors. In this calculation, we split the populace into at least two homogeneous sets. This is done dependent on most huge properties/autonomous factors to make as particular gatherings as could reasonably be expected.

C. KNN (k- Nearest Neighbors)

It very well may be utilized for both order and relapse issues. Be that as it may, it is all the more generally utilized in characterization issues in the business. K nearest neighbors is a straightforward calculation that stores every single accessible case and arranges new cases by a lion's share vote of its k neighbors. The case is to the class is generally normal among its K closest neighbors estimated by a separation work. On the off chance that K = 1, at that point the case is relegated to the class of its closest neighbor. Now and again, picking K ends up being a test while performing KNN displaying.

Conclusion

In this paper, we conclude that the survey report will be just introducing modules of Bitcoin price prediction and machine algorithms. Hear the Comparison table of ML algorithm model accuracy which tells that the linear regression model will have the most accuracy than the other algorithms. In this paper, we conclude that the linear regression algorithm is more efficient than the other algorithms. By taking help from that linear regression algorithm, we can implement the LASSO also. The time complexity reduction in bitcoin price prediction using the LASSO algorithm was tested by referring to all other algorithms and concluded that LASSO is the best among all. The machine learning algorithms will improve that feature idea of cryptocurrencies. That will improve the market price of global investments. In this paper, we proposed a new algorithm to find the feature price accuracy. That helps the customer increments and profits.

References

[1] Bitcoin Price Prediction using Machine Learning, Siddhi Velankar*, Sakshi Valecha*, Shreya Maji* *Department of Electronics & Telecommunication, Pune Institute of Computer Technology, Pune, Maharashtra, Indian 409-415. [2] D. Shah and K. Zhang, ?Bayesian regression and Bitcoin, in 52nd Annual Allerton Conference on Communication, Control, and Computing (Allerton), 2015, pp. [3] Project Based Learning: Predicting Bitcoin Prices using Deep Learning? S. Yogeshwaran; Piyush Maheshwari; Maninder Jeet Kaur; Amity University Dubai Dubai, UAE; IEEE 2019. [4] Predicting the Price of Bitcoin Using Machine Learning Sean McNally; Jason Roche; Simon Caton; Ireland, Dublin, IEEE 2018. [5] Bitcoin Volatility Forecasting with a Glimpse into Buy and Sell Orders? Tian Guo; Albert Bifet; Nino Antulov Fantulin; IEEE 2018. [6] F. Andrade de Oliveira, L. Enrique Zárate and M. de Azevedo Reis; C. NeriNobre, ?The use of artificial neural networks in the analysis and prediction of stock prices, ? in IEEE International Conference on Systems, Man, and Cybernetics, 2011, pp. 2151-2155. [7] M. Daniela and A. BUTOI, ?Data mining on Romania stock market using neural networks for price prediction?. 17,2013. [8] Bitcoin Cost Prediction using Deep Neural Network Technique? Kalpanasonika, Sayasri S, Vinothini, SugaPriya, IEEE 2018. [9] An improved K-Means clustering algorithm?; Juntao Wang, Xiaolong Su, IEEE 2017. [10] Application of Random Forest Algorithm on Feature Subset Selection and Classification and Regression?; Jitendra Kumar Jaiswal, Rita Samikannu IEEE 2017.

Copyright

Copyright © 2023 Dr. Sheetal Phatangare, Piyush Pisal, Bhavik Bondade, Aakash Nihalani. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.