library_name: transformers
language:
- ko
- en
pipeline_tag: text-generation
Mistral-7B-Instruct-v0.2-theme_15k-sft-lora
Model Details
Mistral-7B-Instruct-v0.2-theme_15k-sft-lora is an advanced text generation AI model created by 2digit with a specific focus on news analysis.
Stock-Related Theme Recognition: This model excels in identifying themes and topics relevant to the stock market. It efficiently detects news related to market trends, investment strategies, regulatory changes, and other stock market content. By categorizing articles based on these themes, Mistral-7B-Instruct-v0.2-theme_15k-sft-lora aids analysts and investors in staying updated on pertinent market developments.
Advantages:
- Efficiency: Mistral-7B-Instruct-v0.2-theme_15k-sft-lora streamlines the workflow for news analysis. It automatically extracts critical information from news articles, facilitating quicker insights for financial analysts, journalists, market researchers, and investors.
- Stock Grouping Capability: Leveraging the identified themes, the model enables the grouping of stocks. By categorizing stocks into relevant groups based on extracted themes, Mistral-7B-Instruct-v0.2-theme_15k-sft-lora offers a structured approach to market analysis. This functionality aids in focused analysis, comparison within thematic clusters, and assists in portfolio management and investment strategy formulation.
Overall, Mistral-7B-Instruct-v0.2-theme_15k-sft-lora stands as a robust solution for individuals engaged in news analysis, offering timely insights and facilitating informed decision-making in the dynamic landscape of the stock market.
License
Use of this model requires company approval. Please contact AI@2digit.io. For more details, please refer to the website below: https://2digit.io/#contactus
Dataset
The model was trained on an internal dataset from 2digit, consisting of 15k dataset.
| size | description |
|---|---|
| 15,473 | Human-labeled theme stock dataset |
Evaluation
We measured model accuracy through an internal evaluation system.
| task | accuracy | description |
|---|---|---|
| theme | 0.91 | Extract themes and related companies |