[Relatively] Big Data Processing

I have a table containing approximately 3000 rows and around 10 columns. The data in the table consists mostly of text items belonging to various sets, including items, different categories, and more. My goal is to access this data, conduct semantic analysis, and rephrase the names of the items based on predefined rules, such as name, description, material, and format/size. Specifically, I want to modify the existing names and descriptions based on their relationships with the categories.

However, I want to clarify that GPT-4 cannot read such files (tried via plugins link reader or web pilot) or perform rephrasing tasks. If I manually submit data via screen input, it loses instructions, forgetting tasks, etc.

I perceive GPT-4 as more of a toy than a practical tool in data analysis so far.

I would greatly appreciate any advice or suggestions on effectively accomplishing this task. Please note that I haven’t utilized any programming or similar methods in my approach.

You might need langchain and Vector database as a placeholder to consume such a large data

1 Like