Event sentiment analysis method and device, server and storage medium
A sentiment analysis and event technology, applied in the computer field, can solve problems such as error in sentiment analysis results, inaccurate sentiment analysis results, etc.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0030] figure 1 It is a schematic flowchart of an event sentiment analysis method provided in Embodiment 1 of the present invention. This embodiment is applicable to determining the sentiment analysis result corresponding to each financial news event, and the method can be executed by an event sentiment analysis device , the device can be implemented in the form of software and / or hardware, and optionally, devices such as a PC terminal and a server.
[0031] Such as figure 1 , the method of this embodiment includes:
[0032] S110. Acquire target news data, and process the target news data to determine a target feature vector corresponding to the target news data.
[0033] Wherein, if it is necessary to determine the sentiment analysis result of a certain news data, this news data may be used as the target news data. The target feature vector is the feature vector obtained after processing the target news data as the target feature vector.
[0034] Optionally, the target ne...
Embodiment 2
[0051] On the basis of the above technical solutions, it is also necessary to train a sentiment analysis model. The training and application of the sentiment analysis model are realized based on the pre-built system framework. figure 2 A system architecture diagram for implementing a sentiment analysis model in this embodiment. Such as figure 2 As shown, the system framework includes: AP server, GPU computing device cluster, NAS storage, and database cluster. Among them, at least one AP server is included, and the news data is stored in the NAS; the GPU computing device cluster is used to obtain sample data from the database and train the sentiment analysis model. Data processing based on the above framework can be as follows: During the operation of the system, a preset number of AP servers can read news data from the NAS database to carry out sentiment analysis. After obtaining the sentiment analysis data, the sentiment analysis data can be Send to a preset number of GP...
Embodiment 3
[0075] As a preferred embodiment of the above embodiment, Figure 4 It is a schematic flowchart of an event sentiment analysis method provided by Embodiment 3 of the present invention. This embodiment mainly processes financial event news. Such as Figure 4 As shown, the method includes:
[0076] S410. Acquire the original financial news data, clean the financial news data, and label the original financial news data based on the labeling work to obtain a training sample data set.
[0077] Obtain news data of financial events based on the text crawling server, and obtain market data through fixed quotation form of foreign exchange market, and clean, segment, and label financial news data according to foreign exchange market data, so as to obtain corresponding training for training each sentiment analysis model sample.
[0078] Specifically, when the foreign exchange market is rising (for example, rising more than ten basis points) in a certain period of time, the sample dat...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More