Calculation method of climate news index, computer program and readable storage medium

Abstract

The invention provides a climate news index calculation method, a computer program and a readable storage medium. The method comprises the following steps: acquiring media climate article data; official climate data are obtained, and official climate vocabulary data are extracted from the official climate data; calculating first word frequency data of the official climate vocabulary data in the media climate article data, calculating second word frequency data of the official climate vocabulary data in the official climate data, and calculating inverse document frequency of the media climate article data; multiplying the inverse document frequency by the first word frequency data to obtain a media document vector; multiplying the inverse document frequency by the second word frequency data to obtain an official document vector; calculating the similarity between the official document vector and the media document vector to obtain a climate news index; and outputting the climate news index. The user can use the climate news index to carry out activities such as decision-making and prediction, valuation and pricing related to climate risks.

Description

Technical Field

[0001] The present invention relates to the technical field of climate assessment, and in particular to a calculation method, a computer program and a readable storage medium for a climate news index. Background Art

[0002] Climate change often leads to direct or indirect economic changes, thus creating risks for financial markets. For example, global warming affects the low-carbon policies issued by countries, thus affecting financial activities. Therefore, when predicting the financial market, it is necessary to use climate change indices to study the financial market.

[0003] Changes in financial activities are often related to national climate policies and media climate news. Official policies are usually stable and long-lasting, while media climate news is different every day. If we only study the impact of media climate news on finance, we ignore the impact of official policies and fail to fully estimate climate risks, which leads to people not keeping up with the direction of climate change policy. If we only study the impact of official climate policies on finance, we cannot reflect the impact of climate change on financial markets in the short term.

[0004] There is an existing method for predicting stock price fluctuations based on climate change posts based on machine learning. It uses social media data to predict stock price fluctuations, collects posts through the Internet, classifies and analyzes sentiment based on keyword searches, and uses SHAP to sort influencing factors to improve prediction accuracy. However, this method has limitations: it only relies on social media data, ignores key factors such as official policies and macroeconomics, and is difficult to fully reflect market changes; the quality of social media data has declined, and false information has increased, making the model susceptible to misleading; sentiment analysis relies on keyword searches, and it is difficult to accurately identify complex expressions such as sarcasm and puns, leading to systematic biases; social media data is unstructured, and the model has limited interpretability, making it difficult to provide actionable investment advice; in addition, the data is easy to manipulate, which may lead to market manipulation and prediction distortion. Therefore, the applicability of this method is limited and it is difficult to use it as an independent and reliable stock price prediction tool.

[0005] Moreover, this forecasting method only focuses on the media’s posts about climate change, and does not focus on official policy data, which makes it difficult for the forecast results to fully reflect the real changes in the financial market. In addition, social media discussions often occur after the market has already reacted, rather than being a leading indicator of market changes, which makes this method time-delayed and difficult to capture market trends in a timely manner. Moreover, social media data is collected through emotional posts sent by institutions or individuals, and cannot reflect market trends, making the forecast results distorted. Summary of the invention

[0006] The first object of the present invention is to provide a method for calculating a climate news index that calculates the degree of fit between media information and official guidance.

[0007] The second object of the present invention is to provide a computer device for implementing the above-mentioned method for calculating the climate news index.

[0008] A third object of the present invention is to provide a readable storage medium to which the above-mentioned calculation method of the climate news index is applied.

[0009] To achieve the first purpose of the present invention, the present invention provides a method for calculating a climate news index, which includes: obtaining media climate article data; obtaining official climate data, and extracting official climate vocabulary data from the official climate data; calculating first word frequency data of the official climate vocabulary data in the media climate article data, calculating second word frequency data of the official climate vocabulary data in the official climate data, and calculating the inverse document frequency of the media climate article data; multiplying the inverse document frequency by the first word frequency data to obtain a media document vector; multiplying the inverse document frequency by the second word frequency data to obtain an official document vector; calculating the similarity between the official document vector and the media document vector to obtain a climate news index; and outputting the climate news index.

[0010] As can be seen from the above scheme, by obtaining official climate data and media climate article data, media document vectors and official document vectors are calculated, and the similarity between the two is calculated. The similarity between media document vectors and official document vectors can quantify whether media news pays attention to important climate issues at the official policy level. If the climate news index is high, it proves that the current media news is highly consistent with the official climate policy, reflecting the media's attention to climate risk issues. If the climate news index is low, it proves that the current media's climate news pays little attention to climate issues, and the financial market's mechanism for climate risk is imperfect.

[0011] In a further solution, the media climate article data is set to at least one; the step of obtaining the media climate article data includes: obtaining media article data of at least one date, and extracting each media climate article data from each media article data.

[0012] It can be seen that by obtaining media climate article data every day and calculating the daily climate index, it is possible to reflect whether the climate articles of the day focus on policies, thereby reflecting the flexibility of the index.

[0013] In a further solution, one media climate article data corresponds to a first word frequency data, and one media climate article data also corresponds to a climate index.

[0014] It can be seen that by obtaining media climate article data every day, the daily climate index can be calculated.

[0015] In a further solution, the step of obtaining media article data of at least one date includes: obtaining media article data of at least one date of the first media.

[0016] It can be seen that media article data are obtained through a single media (e.g. People’s Daily) and the climate news index of that media is calculated to reflect the media’s attention to climate issues, which is based on official policies, thus reducing the impact of the mixing of data from multiple media on the climate news index, which can lead to bias.

[0017] In a further solution, after obtaining the climate news index, the following steps are further performed: obtaining media article data of at least one date of the second media, and calculating the climate news index corresponding to the media article of the second media.

[0018] It can be seen that different media have different political, economic and social backgrounds. By calculating the climate news index of each media, it is helpful to fully grasp the media's attention to climate and provide users with multi-angle reference.

[0019] In a further solution, after obtaining the media climate article data, the following steps are also performed: preprocessing the media climate article data; the preprocessing step includes: removing punctuation information and stop word information of the media climate article data; and performing word segmentation processing on the media climate article data.

[0020] It can be seen that by preprocessing media climate article data, the interference of useless information on subsequent processing can be reduced.

[0021] In a further solution, after outputting the climate news index, the following is further performed: determining whether the scheduled update time has arrived, and if so, obtaining the climate article data of the latest date and calculating the latest climate news index.

[0022] It can be seen that by setting an update time, the Climate News Index can be updated and thus maintain its dynamic nature.

[0023] In a further solution, the step of calculating the similarity between the official document vector and the media document vector includes: calculating the cosine similarity between the official document vector and the media document vector.

[0024] It can be seen that cosine similarity can effectively measure the similarity of text content and is suitable for high-dimensional text data.

[0025] To achieve the second objective of the present invention, the present invention provides a computer device, including a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed, the above-mentioned method for calculating the climate news index can be implemented.

[0026] To achieve the third objective of the present invention, the present invention provides a readable storage medium having a computer program stored thereon, which can implement the above-mentioned method for calculating the climate news index when the computer program is executed. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flow chart of an embodiment of a method for calculating a climate news index of the present invention.

[0028] The present invention is further described below in conjunction with the accompanying drawings and embodiments. DETAILED DESCRIPTION

[0029] The present invention provides a method for calculating a climate news index, which measures media news data and official document data, so that the calculated climate news index can reflect the current media news attention to climate issues. It is based on official policy data, so that users can dynamically adjust hedging strategies according to the climate news index.

[0030] Example of calculation method of Climate News Index: See also Figure 1 The calculation method of the climate news index in this embodiment is executed by a computer device. First, step S11 is executed to obtain media climate article data. The media climate article data is set as data of at least one date. Specifically, the media climate article data is obtained from media article data. The media article data is the daily news of the first media, and the media climate article data is the media article data of the first media on at least one date, that is, the climate article in the daily news of the first media. For example, the daily news of the People's Daily is obtained, and the daily climate article is obtained therefrom.

[0031] After obtaining the media climate article data, the media climate article data is also preprocessed. The specific preprocessing steps include: removing punctuation information and stop word information from the media climate article data, performing word segmentation on the media climate article data, etc. The media climate article data is preprocessed using natural language processing technology. The media climate article data is preprocessed so that it is not disturbed by cluttered information during the subsequent climate news index calculation, ensuring the accuracy of data quality.

[0032] After obtaining media climate article data, execute step S12 to obtain official climate data and extract official climate vocabulary data from the official climate data. The official climate data is policy climate data, and the official climate data is a white paper on climate change collected by official authoritative sources. The official climate vocabulary data includes common climate change terms, such as greenhouse effect, carbon footprint, and climate crisis. After extracting the official climate vocabulary data, count the frequency of the official climate vocabulary data in the official climate data, set a frequency threshold to filter the official climate vocabulary in the official climate vocabulary data, and delete the official climate vocabulary below the frequency threshold from the official climate vocabulary data to ensure that the constructed vocabulary is representative and effective.

[0033] After extracting the official climate vocabulary data, step S13 is executed to calculate the first word frequency data, which is the word frequency of the official climate vocabulary data in a media climate article data. One media climate article data corresponds to one word frequency data. The calculation formula of the first word frequency data is: , where TF(t,d) is the first word frequency data, t is the number of times the official climate vocabulary data appears in the media climate article data, and d1 is the total number of words in the media climate article data.

[0034] After the first word frequency data is calculated, step S14 is executed to calculate the second word frequency data of the official climate vocabulary data in the official climate data. The calculation method of the second word frequency data is the same as the calculation method of the first word frequency data.

[0035] After the second word frequency data is calculated, step S15 is executed to calculate the inverse document frequency of the media climate article data. The calculation formula for the inverse document frequency of the media climate article data is: Wherein, IDF(t) is the inverse document frequency, n is the total number of media climate article data of the first media, and d2 is the number of media climate article data containing official climate vocabulary data.

[0036] After calculating the inverse document frequency of the media climate article data, step S16 is executed to calculate the media document vector and the official document vector. The media document vector is obtained by multiplying the inverse document frequency with the first word frequency data, and the official document vector is obtained by multiplying the inverse document frequency with the second word frequency data. One first word frequency data corresponds to one media document vector.

[0037] The official document vector is obtained by multiplying the inverse document frequency of the media document vector by the second word frequency data, which can reflect the impact of the vocabulary usage in the media climate news data on the similarity calculation. If the inverse document frequency of the official climate data is directly calculated, the two unprocessed inverse document frequencies can be directly compared, which will lead to some common words in the media climate article data but less in the official climate data being over-underestimated.

[0038] After the calculation of the media document vector and the official document vector is completed, step S17 is executed to calculate the similarity between the official document vector and the media document vector to obtain the climate news index. The similarity is the cosine similarity, and its calculation formula is: Among them, cosine similarity(A,B) is the cosine similarity between the official document vector and the media document vector, A is the official document vector, and B is the media document vector. One media climate article data corresponds to one climate news index.

[0039] The similarity between the media document vector and the official document vector can quantify whether the media news pays attention to important climate issues at the official policy level. If the climate news index is high, it proves that the current media news is highly consistent with the official climate policy, reflecting the media's attention to core climate risks. If the climate news index is low, it proves that the current media's climate news pays less attention to policies. Users can observe the climate news index to understand whether the media pays attention to policies, so as to dynamically adjust hedging strategies. And the climate news index combines the media document vector and the official document vector, and the index is dynamic. This method can quantify whether the media effectively conveys the official and authoritative information on climate change to the public and the market, and also provides data support for investors to dynamically adjust their hedging strategies for climate risks.

[0040] The application of the Climate News Index covers many fields, including financial markets, corporate decision-making, economic forecasts, the insurance industry, policy supervision, media and information analysis, the application of artificial intelligence, and the promotion of international climate negotiations and cooperation.

[0041] After calculating the climate news index, the climate news index is output. The output climate news index can be output on a display connected to the processor, or can be a front-end page of a website. After the climate news index is displayed on the front-end page, a preset update time is also set. The preset update time can be 0:00 every day. It is determined whether the preset update time has been reached. If so, the media article data of the latest date is obtained to calculate the latest climate news index.

[0042] After the climate news index is calculated, the climate news index is the climate news index of the first media. After the climate news index of the first media is calculated, media article data of at least one date of the second media is obtained, and the climate news index corresponding to the media article data of the second media is calculated. For example, the first media is the People's Daily and the second media is the Economic Daily. Calculate the climate news index of different daily newspapers according to different needs. Different media have different political, economic and social backgrounds. By calculating different climate news indexes of different media, it is helpful to fully grasp the attention of different media news on climate issues based on official policy data, and provide users with different references. The calculation method of the climate news index in this embodiment uses media article data of official media, such as People's Daily and Economic Daily, and is more accurate when predicting market directions.

[0043] Computer device embodiment: The computer device of this embodiment includes a processor and a memory. The memory stores a computer program. When the processor executes the computer program, the above-mentioned method for calculating the climate news index is implemented.

[0044] The computer device may include but is not limited to a processor and a memory. Those skilled in the art will appreciate that the computer device may include more or fewer components, or a combination of certain components, or different components, for example, the computer device may also include input and output devices, network access devices, buses, etc.

[0045] Readable storage medium embodiment: The method for calculating the climate news index in the computer device described in the above embodiment can be stored in a readable storage medium in the form of a computer program. When the computer program is executed by a processor, the steps of the embodiment of the method for calculating the climate news index in the computer device can be completed. The computer readable medium can be a computer readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0046] The above are only preferred embodiments of the present invention, but the design concept of the invention is not limited thereto. Without departing from the concept of the present invention, more other equivalent embodiments may be included. It is obvious to those skilled in the art that various changes, readjustments and substitutions can be made without departing from the protection scope of the present invention.

Claims

1. The calculation method of the climate news index is characterized by: The method includes: Obtain media climate article data; obtaining official climate data, and extracting official climate vocabulary data from the official climate data; Calculate the first word frequency data of the official climate vocabulary data in the media climate article data, calculate the second word frequency data of the official climate vocabulary data in the official climate data, and calculate the inverse document frequency of the media climate article data; Multiplying the inverse document frequency by the first word frequency data to obtain a media document vector; Multiplying the inverse document frequency by the second word frequency data to obtain an official document vector; Calculating the similarity between the official document vector and the media document vector to obtain a climate news index; Outputs the climate news index.

2. The method for calculating the climate news index according to claim 1, characterized in that: The steps to obtain media climate article data include: Acquire media article data of at least one date, and extract each media climate article data from each media article data.

3. The method for calculating the climate news index according to claim 2, characterized in that: One piece of media climate article data corresponds to one piece of the first word frequency data, and one piece of media climate article data also corresponds to one piece of the climate news index.

4. The method for calculating the climate news index according to claim 2, characterized in that: The steps for obtaining media article data for at least one date include: Acquire media article data of at least one date of a first media.

5. The method for calculating the climate news index according to any one of claims 1 to 4, characterized in that: After getting the climate news index, also execute: Acquire media article data of at least one date of a second media, and calculate a climate news index corresponding to the media article data of the second media.

6. The method for calculating the climate news index according to any one of claims 1 to 4, characterized in that: After obtaining the media climate article data, the following is also performed: Preprocessing the media climate article data; The step of preprocessing the media climate article data includes: Removing punctuation information and stop word information from the media climate article data; The media climate article data is subjected to word segmentation processing.

7. The method for calculating the climate news index according to claim 6, characterized in that: After outputting the climate news index, also execute: Determine whether the preset update time has been reached. If so, obtain media article data of the latest date and calculate the latest climate news index.

8. The method for calculating the climate news index according to any one of claims 1 to 4, characterized in that: The step of calculating the similarity between the official document vector and the media document vector comprises: The cosine similarity between the official document vector and the media document vector is calculated.

9. A computer device, characterized in that The invention comprises a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed, the method for calculating the climate news index according to any one of claims 1 to 8 is implemented.

10. A readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed, the method for calculating the climate news index according to any one of claims 1 to 8 is implemented.

Images