File size: 1,709 Bytes
ebeaa49
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
{
  "model_type": "custom",
  "model_name": "Product-Review-Grader",
  "description": "A model that analyzes product reviews, generates summaries, extracts pros and cons, and assigns grades based on sentiment.",
  "schema": {
    "preprocessing": {
      "stopwords": [
        "the", "is", "in", "at", "on", "a", "an", "and", "or", "for", "to", "of", "with", "that", "by", "it"
      ],
      "text_cleaning": "removes numbers, punctuation, and extra spaces, converts to lowercase"
    },
    "task": {
      "name": "sentiment_analysis_and_grading",
      "description": "Generates sentiment analysis and grades for product reviews using a fine-tuned Gemini model.",
      "input": {
        "review_text": "A string representing a product review."
      },
      "output": {
        "sentiment_label": ["POSITIVE", "NEUTRAL", "NEGATIVE"],
        "grade": "A float number from 1 to 5 representing the grade."
      }
    },
    "summarization": {
      "input": "Text from multiple reviews combined into a single string.",
      "output": "A summarized string of the main points of the reviews."
    },
    "pros_and_cons_extraction": {
      "input": "Text from multiple reviews combined into a single string.",
      "output": {
        "pros": "A string with pros extracted from the text.",
        "cons": "A string with cons extracted from the text."
      }
    },
    "parameters": {
      "min_raters": 35,
      "weighting_formula": "Weighted average combining product rating and global average rating."
    }
  },
  "dependencies": {
    "libraries": [
      "streamlit",
      "pandas",
      "google-generativeai",
      "os",
      "json",
      "re"
    ],
    "python_version": "3.10"
  }
}