Research on Data Mining and Investment Recommendation of Individual Users Based on Financial Time Series Analysis

Introduction

The computerization of society has significantly enhanced the ability of people to generate and collect data. Huge amounts of data are pouring out from every corner of our lives. The explosive growth of stored or transient data has spurred the need for new technologies and automated tools to help us intelligently convert massive amounts of data into useful information and knowledge, leading to the emergence of a frontier discipline in computer science called data mining. Data mining refers to the automatic extraction of unknown but valuable representative knowledge from massive, uncertain, incomplete, fuzzy and noisy data sets. These patterns are hidden in large databases, data warehouses, other large amounts of information base or data stream. Data mining can be regarded as the process of discovering the existing relationships and rules, association patterns and trend prediction from massive data sets (Qua et al., 2013). Data mining can discover potential patterns and find out individual user information ignored by enterprise decision makers, so that it can be easily understood and clearly reflected to decision makers and provide reference for enterprise decision-making. Data mining is the mining of data with potential value of information, and this information has implicit, previously unknown, non-trivial, meaningful features. Data mining is a high-level process for identifying effective, novel, potentially useful and ultimately understandable patterns from data sets. It includes data cleaning, data integration, data selection and transformation, data mining, pattern evaluation and knowledge representation, and applies various methods to discover implicit rules and patterns from data sequences (Durante et al., 2014).

In the financial market, information continuously affects the market price changes. It is important and difficult to make the price trend judgment quickly. In practical application, it is often necessary to find the possible correlations between different financial time series. This kind of relationship has important reference value for people to understand the interaction of various financial time series more thoroughly and make reasonable decisions accordingly (Adhikari & Agrawal, 2014). Time series analysis has become an indispensable part of financial market research and is one of the important methods of financial quantitative analysis. Many research results in financial markets are based on time series analysis (Gui et al., 2015). Up to now, the importance of time series analysis has been widely recognized in the world. Financial market is a huge system, which is affected by many factors, and has a very complex movement law. Time series data is its comprehensive external manifestation. Because the data in the financial market is mainly time series, financial market analysis is often called financial time series analysis. Financial time series is the real measurement and record of the development data of financial variables in reality, and then through the quantitative analysis and summary of the corresponding mathematical statistics methods, we can find the inherent laws and behavioral characteristics of the financial market, thus providing important judgment basis for investment decision-makers. It is the theoretical basis of capital asset pricing and financial risk prevention.