CN-122022874-A - Predictive analysis system and method

Abstract

The musical revenue prediction analysis system may include a memory and a processor for storing instructions. The instructions, when executed by the processor, may be controlled by the system to collect a plurality of data including basic information, economic indicators, leading indicators, and SNS data of the sound source, perform preprocessing including data normalization, missing value processing, and outlier detection on the plurality of data, predict future benefits of the sound source from the preprocessed data through deep learning analysis, and generate confidence intervals and sensitivity analysis results for the predicted future benefits.

Inventors

• Pu Xuyan

Assignees

• 朴㥠沿

Dates

Publication Date

20260512

Application Date

20241125

Priority Date

20241112

Claims (1)

1. 1. A predictive analysis system, which is a musical revenue predictive analysis system, comprising: memory, store instructions, and The processor may be configured to perform the steps of, When executed by a processor, the instructions include: collecting multiple data including basic information of sound source, economic index, leading index and SNS data, Performing preprocessing on the plurality of data, including data normalization, missing value processing and outlier detection, Predicting future copyright charges of the sound source from the preprocessed data through deep learning analysis, Calculating the basic value of the copyright according to the predicted copyright expense, The final investment value is calculated through macroscopic economic index and investment emotion analysis, The predictive analysis system is used for controlling the generation of confidence intervals and sensitivity analyses of the predicted investment values.

Description

Predictive analysis system and method Technical Field The invention relates to a system and a method for predicting and analyzing the investment value of a music copyright by utilizing an artificial intelligence technology, in particular to a system and a method for predicting and analyzing the final investment value of the music copyright by predicting future copyright cost through audio related data and calculating the basic value of the copyright according to the future copyright cost, and further reflecting the investment environment and the investment emotion by combining a macroscopic economic index. Background In recent years, copyright investment has become a new investment tool for music markets. Investors can invest in music copyrights and gain royalty revenues such as equity, but lack systematic analysis tools for proper investment decisions. Existing music evaluation systems are limited to simply using past leaderboard ranking and copyright revenue information, or analyzing SNS data. In addition, although the music market is closely related to the overall financial market and economic status, it is difficult to accurately predict profits since these macroscopic factors are not considered. Heretofore, korean patent No. 10-2007-0046415, "music stock trading system using a mobile communication terminal and method thereof," discloses a system and method for downloading music and trading the music as a stock using the communication terminal. If a user wants to invest in a certain music source, it includes only one configuration that allows the user to download and trade music as a stock, the user or investor must decide which music source is good in prospect or which can provide stable revenue, and there is still a problem in that the investor cannot help them make investment decisions for the music source. Prior art literature Patent literature Patent document 1 Korean patent No. 10-2007-0046415 Disclosure of Invention Problems to be solved by the invention The present invention aims to address the limitations of such prior art by enabling more accurate profit predictions by integrating data relating to sound sources with macro economic indicators. The present invention uses deep learning techniques to learn correlations between complex data and to improve prediction accuracy. The present invention aims to detect future revenue changes at an early stage by using leading indicators such as an online search amount and a YouTube browsing amount. Means for solving the problems The sound source revenue prediction analysis system may include a memory and a processor to store instructions. The instructions, when executed by the processor, collect a plurality of data including sound source base information, economic indicators, lead indicators, and SNS data, and perform preprocessing including data normalization, missing value processing, and anomaly detection on the plurality of data, and predict future revenues of sound sources from the preprocessed data through deep learning analysis, which can be controlled to generate confidence intervals and sensitivity analysis for predicting future returns. The musical revenue prediction analysis system has the ability to analyze future revenue of a musical source through complex data processing and predictive models. The system includes a memory storing instructions and a processor executing the instructions and may perform a series of processes to collect and analyze data associated with various sound sources. The system 400 first collects basic information of the sound source. This includes the title, genre, release date, composer and artist information, and artist profile. These basic information provide an important basis for determining the characteristics and popularity estimate of the sound source. For example, from past trends, new songs of popular genres or specific artists can be predicted to have higher revenue potential. The system also collects economic indicators and reflects them in the prediction of musical revenue. The economic index comprises macroscopic data such as consumption expenditure level, economic growth rate, loss rate, and expansion rate of the general market. For example, when the overall economy is good, people are more likely to spend more entertainment, while when the economy is low, they may reduce purchasing music or subscribing to streaming media. Thus, the economic indicators can be used as important external variables in revenue prediction for music sources. Leading indicators are also collected, which are used to predict trends or potential revenue fluctuations within the music market. For example, if a genre is temporarily popular, or an artist is in focus, revenue predictions for music similar to that genre or artist may be positively impacted. These lead indicators reflect recent trends in music consumption or streaming media patterns so that more accurate predictions can be made. The system 400 may collect various external