File size: 6,885 Bytes
3852d49
3683f9c
 
3852d49
3683f9c
2162f38
3683f9c
 
3852d49
 
3683f9c
2162f38
3683f9c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2162f38
3683f9c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2162f38
 
3852d49
3938c4d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d2f3083
3852d49
d2f3083
3852d49
 
 
3938c4d
3852d49
3938c4d
2162f38
 
 
d2f3083
 
 
2162f38
d2f3083
 
2162f38
 
 
 
d2f3083
 
 
 
 
3938c4d
d2f3083
2162f38
d2f3083
 
3938c4d
3852d49
d2f3083
3938c4d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d2f3083
 
 
3938c4d
 
 
 
 
 
 
 
d2f3083
3852d49
 
3683f9c
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
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
import gradio as gr
import pandas as pd
import re
from huggingface_hub import InferenceClient
import plotly.express as px
from collections import Counter

# Initialize Hugging Face client
client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")

def parse_message(message):
    """Extract information from a chat message using regex."""
    info = {}
    
    # Extract timestamp and phone number
    timestamp_match = re.search(r'\[(.*?)\]', message)
    phone_match = re.search(r'\] (.*?):', message)
    
    if timestamp_match and phone_match:
        info['timestamp'] = timestamp_match.group(1)
        info['phone'] = phone_match.group(1)
        
        # Extract rest of the message
        content = message.split(':', 1)[1].strip()
        
        # Extract name
        name_match = re.match(r'^([^•\n-]+)', content)
        if name_match:
            info['name'] = name_match.group(1).strip()
            
        # Extract affiliation
        affiliation_match = re.search(r'[Aa]ffiliation:?\s*([^•\n]+)', content)
        if affiliation_match:
            info['affiliation'] = affiliation_match.group(1).strip()
            
        # Extract research field/interests
        field_match = re.search(r'([Ff]ield of [Ii]nterest|[Dd]omaine de recherche|[Rr]esearch area|[Aa]reas of interest):?\s*([^•\n]+)', content)
        if field_match:
            info['research_field'] = field_match.group(2).strip()
            
        # Extract thesis topic
        thesis_match = re.search(r'[Tt]hesis:?\s*([^•\n]+)', content)
        if thesis_match:
            info['thesis_topic'] = thesis_match.group(1).strip()
    
    return info

def create_researcher_df(chat_history):
    """Convert chat messages to structured DataFrame."""
    researchers = []
    messages = chat_history.split('\n')
    
    for message in messages:
        if message.strip():
            info = parse_message(message)
            if info:
                researchers.append(info)
    
    df = pd.DataFrame(researchers)
    return df

def analyze_research_fields(df):
    """Analyze and categorize research fields."""
    if 'research_field' not in df.columns:
        return pd.Series()
        
    fields = df['research_field'].dropna()
    # Split fields and flatten
    all_fields = [field.strip().lower() for fields_list in fields for field in fields_list.split(',')]
    return pd.Series(Counter(all_fields))

def create_visualizations(df):
    """Create visualizations from the researcher data."""
    figures = []
    
    # 1. Affiliation Distribution
    if 'affiliation' in df.columns and not df['affiliation'].empty:
        affiliation_counts = df['affiliation'].value_counts()
        fig_affiliation = px.pie(
            values=affiliation_counts.values,
            names=affiliation_counts.index,
            title='Distribution of Researchers by Affiliation'
        )
        figures.append(fig_affiliation)
    
    # 2. Research Fields Analysis
    field_counts = analyze_research_fields(df)
    if not field_counts.empty:
        fig_fields = px.bar(
            x=field_counts.index,
            y=field_counts.values,
            title='Popular Research Fields',
            labels={'x': 'Field', 'y': 'Count'}
        )
        figures.append(fig_fields)
        
    return figures[0] if figures else None

def analyze_chat_history(chat_history_text):
    """Analyze chat history and return DataFrame, plot, and summary."""
    if not chat_history_text.strip():
        return None, None, "No chat history provided."
    
    df = create_researcher_df(chat_history_text)
    
    if df.empty:
        return None, None, "No data could be extracted from the chat history."
    
    # Generate analysis summary
    summary = f"Analysis of {len(df)} researchers:\n"
    if 'affiliation' in df.columns:
        summary += f"- Institutions represented: {df['affiliation'].nunique()}\n"
    
    field_counts = analyze_research_fields(df)
    if not field_counts.empty:
        top_fields = field_counts.nlargest(3)
        summary += "- Top research fields:\n"
        for field, count in top_fields.items():
            summary += f"  • {field}: {count} researchers\n"
    
    # Create visualization
    fig = create_visualizations(df)
    
    return df, fig, summary

def process_message(
    message,
    chat_history,
    system_message,
    max_tokens,
    temperature,
    top_p
):
    """Process message and return response."""
    try:
        # Generate response using the LLM
        messages = [{"role": "system", "content": system_message}]
        for user_msg, bot_msg in chat_history:
            messages.append({"role": "user", "content": user_msg})
            messages.append({"role": "assistant", "content": bot_msg})
        messages.append({"role": "user", "content": message})

        response = client.chat_completion(
            messages,
            max_tokens=max_tokens,
            temperature=temperature,
            top_p=top_p,
        )
        
        bot_message = response.choices[0].message.content
        chat_history.append((message, bot_message))
        
        return chat_history

    except Exception as e:
        error_message = f"Error: {str(e)}"
        chat_history.append((message, error_message))
        return chat_history

with gr.Blocks(title="CohortBot") as demo:
    with gr.Row():
        with gr.Column(scale=2):
            chatbot = gr.Chatbot(label="Chat History")
            msg = gr.Textbox(label="Message", placeholder="Type your message here...")
            with gr.Row():
                system_msg = gr.Textbox(value="You are a friendly Research Community Chatbot.", label="System message")
            with gr.Row():
                max_tokens = gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens")
                temperature = gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature")
                top_p = gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p")
        
        with gr.Column(scale=1):
            chat_history_text = gr.Textbox(label="Chat History for Analysis", lines=10)
            analyze_btn = gr.Button("Analyze Chat History", variant="primary")
            with gr.Row():
                analysis_text = gr.Textbox(label="Analysis Summary", lines=4)
            with gr.Row():
                researcher_table = gr.Dataframe(label="Extracted Researcher Data")
            with gr.Row():
                plot = gr.Plot(label="Community Analysis")
    
    msg.submit(
        process_message,
        [msg, chatbot, system_msg, max_tokens, temperature, top_p],
        [chatbot]
    )
    
    analyze_btn.click(
        analyze_chat_history,
        inputs=[chat_history_text],
        outputs=[researcher_table, plot, analysis_text]
    )

if __name__ == "__main__":
    demo.launch()