CRYPTOCURRENCY PRICE PREDICTION USING DATA SCIENCE

Abstract

After the boom and bust of blockchain in recent years, cryptocurrency has been increasingly regarded as an investment asset. Because of its highly volatile nature, there is a need for good predictions on which to base investment decisions. Although existing studies have leveraged machine learning for more accurate cryptocurrency price prediction, few have focused on the feasibility of applying different modelling techniques to samples with different data structures and dimensional features. To predict cryptocurrency price at different frequencies using machine learning techniques, we first classify cryptocurrency price by daily price and high-frequency price. A set of high-dimension features including property and network, trading and market, attention and gold spot price are used for cryptocurrency daily price prediction, while the basic trading features acquired from a cryptocurrency exchange are used for 5-minute interval price prediction. Price control by a number of organizations has had a significant impact on the level of one main or central control over them, affecting relationships with other businesses and international trade. Furthermore, the ever-changing oscillations suggest a more accurate means of projecting this price is desperately needed. Thus, using deep learning techniques such as the recurrent neural network (RNN) and the long short-term memory (LSTM), gated recurrent unit (GRU), which are effective learning models for training data, we must design a method for the accurate prediction of by considering various factors such as market cap, maximum supply and, volume, circulating supply. The proposed method is written in Python and tested on benchmark datasets. The results show that the proposed method can be used to make reliable predictions. Thus, the neural network, which has been used by academics in numerous fields over the past ten years as one of the intelligent data mining tools

Specification

Description:FIELD OF INVENTION
The proposed system is using data science technology to predict the crypto currency price. Here the system will predict crypto currency price with help 1h, 24h, 7days correlations, sentimental analysis of recent news of the particular coin a with help of live data, the model will predict the price trend (up-trend, down-trend) and here we will recommend the coin based on the analysis of that particular coin. In our system we will get live data from coin Api. Those live data will be preprocessed with help python pandas library and then passing those data to model. Here all those data will be pre trained and forecasted. Those forecasted data will be stored into database to and rendered quickly. The new system will use Long Short Term Memory networks - usually just called "LSTMs" - are a special kind of RNN, capable of learning long-term dependencies. A recurrent neural network (RNN) is a class of artificial neural networks where connections between nodes form a directed graph along a temporal sequence. This allows it to exhibit temporal dynamic behavior The proposed system is uses sentimental analysis that will help user to know more details about crypto coin news. The sentimental analysis will classify the news and show the positive, neutral, Negative news of that particular coin. Here we will recommend the coins based on the sentimental analysis if positive means BUY, negative means SELL, Neutral means HOLD. This system will show the current trend of the coin with help of animated gif images it will show bull if the current trend is positive and bear if current is negative. Then it will show the trend graph with help of google-chart. This system is also showing daily top gainer and looser that help user to understand the most effective and ineffective coin of previous day. Then live data is used to predict the trend of the Coins.
BACKGROUND OF THE INVENTION
The financial sector has already been noted as a potential use case for on-chain data. Comparing blockchain networks to companies, there are various similarities to be observed - most networks have a coin, equivalent to the coins, and both have a price and are traded on exchanges. There are fundamental differences: blockchain networks are not structured like companies, there is no need for executives, and the economic parameters are written in code. The most recent story about Bitcoin is that it is an improved version of Gold for its store of value and non-inflationary properties, coupled with its lower friction of moving assets around. However, seeing as their popularity as financial instruments are increasing and ever more credible investors are adding them to their portfolios [44], we can ascertain that cryptocurrencies have gained their category as crypto-assets. They are traded on dedicated exchanges and are becoming significant parts of investors' portfolios around the world.
Hence, the need arises for tools and protocols to analyse these crypto-assets as financial instruments and decide whether to make them part of one's portfolio or not. On-chain metrics provide the best basis for such tools, as decentralized and tamper-proof data that enables anyone access to in-depth information about a network's usage, with no entity having the ability to censor it. We view this as akin to everybody having access to insider information on any publicly traded company. We consider this could act as a way of levelling the financial playing field and getting us closer to the efficient market hypothesis [45] which states that all the publicly and privately accessible information available to market participants is already reflected in the price
SUMMARY OF THE INVENTION
In this invention, we propose a model for the prediction of coin price.
Fig 1.1 Architecture of the Invention
The RNN (LSTM) deep learning Model will train and forecast coin price. Those forecasted data will be stored in mongo database with help of mongo library. Here the client will be requesting the data with help of Coin name in coin API. The API will be creating request to similar APIs such as Forecasted API, Sentimental Analysis API and Financial Ratios API, Then the system will fetch all necessary data. After fetched those data will be preprocessed with help of python panda's library. Then all the data will be forwarded to necessary API, with help of Django the data will to rendered quickly and shown to user. Coin API will be giving all live price of coins. Here we will be fetching all live data and shown to user in a component format if the value is down-trend it will show red color, if the value is up-trend it will show green color.
Hardware and Software Requirements
Hardware requirements:
 Intel core i3 Processor
 2.0 GHz or Above Clock Speed
 8 GB and Above RAM
 250 GB Hard Disks
 Internet Connection
Software requirements:
 Windows 10 OS
 Python, Django
 Visual Studio Code
 Mongo DB
 Chrome or Firefox browser
We have succeeded in our aim to develop a system that can be used to forecast the crypto coin price with help of historical data and here three techniques have been utilized in this paper: Forecast data, Sentimental Analysis, on the coin api dataset. All the techniques have shown an improvement in the accuracy of predictions, thereby yielding positive results. Use of recently introduced deep learning techniques in the prediction of coin have yielded promising results and thereby marked the use of them in profitable exchange schemes. It has led to the conclusion that it is possible to predict currency market with more accuracy and efficiency using deep learning techniques.
Fig 1.2 Price Prediction , Claims:1.The system can be used to forecast the crypto coin price with help of historical data and here three techniques have been utilized in this paper:
2. The device can Forecast data, Sentimental Analysis, on the coin api dataset. All the techniques have shown an improvement in the accuracy of predictions, thereby yielding positive results.
3.The Use of these device recently introduced deep learning techniques in the prediction of coin have yielded promising results and thereby marked the use of them in profitable exchange schemes.
4. It has led to the conclusion that it is possible to predict currency market with more accuracy and efficiency using deep learning techniques.

Documents