File size: 11,085 Bytes
b276dbe
159fe5e
b276dbe
 
 
 
 
 
 
 
 
 
2d6d126
b276dbe
f45ad41
9e91c10
b276dbe
 
 
 
2d6d126
cf45911
2d6d126
cf45911
 
2d6d126
 
 
 
 
cf45911
 
2d6d126
cf45911
 
 
 
2d6d126
 
cf45911
 
2d6d126
 
 
 
 
 
 
159fe5e
6d5eed2
 
 
 
 
 
 
 
cf45911
2d6d126
 
159fe5e
 
 
2d6d126
 
 
 
159fe5e
2d6d126
 
 
 
 
 
 
 
 
 
 
 
 
 
159fe5e
 
 
 
 
 
 
cf45911
2d6d126
 
 
 
 
 
 
 
cf45911
 
b276dbe
2d6d126
 
b276dbe
 
 
 
9e91c10
 
b276dbe
 
 
 
 
 
 
 
 
 
 
1c1d54b
b276dbe
 
 
 
 
 
 
 
cf45911
b276dbe
cf45911
b276dbe
cf45911
b276dbe
cf45911
 
b276dbe
 
 
 
 
 
 
9e91c10
b276dbe
 
 
1c1d54b
 
b276dbe
 
 
 
 
 
9f87bf2
b276dbe
9e91c10
b276dbe
 
 
 
 
 
 
 
 
 
 
9e91c10
 
 
b276dbe
 
 
 
 
 
 
 
2d6d126
b276dbe
2d6d126
 
b276dbe
159fe5e
2d6d126
b276dbe
 
2d6d126
e12f791
2d6d126
 
 
842db18
2d6d126
 
b276dbe
b893da2
e12f791
 
 
842db18
e12f791
2d6d126
 
 
 
 
 
 
 
 
 
 
 
842db18
2d6d126
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9e91c10
2d6d126
 
 
9e91c10
159fe5e
2d6d126
 
 
 
 
 
 
9e91c10
2d6d126
 
 
 
 
 
 
9e91c10
2d6d126
 
 
 
 
9e91c10
2d6d126
9e91c10
b276dbe
2d6d126
 
 
 
 
 
b276dbe
1c1d54b
b276dbe
159fe5e
 
 
 
 
 
 
 
 
 
 
b276dbe
2d6d126
 
 
159fe5e
b276dbe
2d6d126
159fe5e
2d6d126
 
 
 
 
 
 
 
b276dbe
 
 
 
d5a9a0d
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
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
import streamlit as st
import datetime
import pickle
import numpy as np
import rdflib
import torch
import os
import requests
from rdflib import Graph as RDFGraph, Namespace
from sentence_transformers import SentenceTransformer
from dotenv import load_dotenv

# === CONFIGURATION ===
load_dotenv()
ENDPOINT_URL = "https://api-inference.huggingface.co/models/mistralai/Mistral-7B-Instruct-v0.3"
HF_API_TOKEN = os.getenv("HF_API_TOKEN")
EMBEDDING_MODEL = "intfloat/multilingual-e5-base"
DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
EX = Namespace("http://example.org/lang/")

# === STREAMLIT UI CONFIG ===
st.set_page_config(
    page_title="Language Atlas: South American Indigenous Languages",
    page_icon="๐ŸŒ",
    layout="wide",
    initial_sidebar_state="expanded",
    menu_items={
        'About': "## AI-powered analysis of endangered indigenous languages\n"
                "Developed by Departamento Acadรฉmico de Humanidades"
    }
)

# === CUSTOM CSS ===
st.markdown("""
<style>
    .header {
        color: #2c3e50;
        border-bottom: 2px solid #4f46e5;
        padding-bottom: 0.5rem;
        margin-bottom: 1.5rem;
    }
    .feature-card {
        background-color: #f8fafc;
        border-radius: 8px;
        padding: 1rem;
        margin: 0.5rem 0;
        border-left: 3px solid #4f46e5;
    }
    .response-card {
    background-color: #fdfdfd;
    color: #1f2937;
    border-radius: 8px;
    padding: 1.5rem;
    box-shadow: 0 2px 6px rgba(0,0,0,0.08);
    margin: 1rem 0;
    font-size: 1rem;
    line-height: 1.5;
    }
    .language-card {
        background-color: #f9fafb;
        border-radius: 8px;
        padding: 1rem;
        margin: 0.5rem 0;
        border: 1px solid #e5e7eb;
    }
    .sidebar-section {
        margin-bottom: 1.5rem;
    }
    .sidebar-title {
        font-weight: 600;
        color: #4f46e5;
    }
    .suggested-question {
        padding: 0.5rem;
        margin: 0.25rem 0;
        border-radius: 4px;
        cursor: pointer;
        transition: all 0.2s;
    }
    .suggested-question:hover {
        background-color: #f1f5f9;
    }
    .metric-badge {
        display: inline-block;
        background-color: #e8f4fc;
        padding: 0.25rem 0.5rem;
        border-radius: 4px;
        font-size: 0.85rem;
        margin-right: 0.5rem;
    }
    .tech-badge {
        background-color: #ecfdf5;
        color: #065f46;
        padding: 0.25rem 0.5rem;
        border-radius: 4px;
        font-size: 0.75rem;
        font-weight: 500;
    }
</style>
""", unsafe_allow_html=True)

# === CORE FUNCTIONS ===
@st.cache_resource(show_spinner="Loading AI models and knowledge graphs...")
def load_all_components():
    embedder = SentenceTransformer(EMBEDDING_MODEL, device=DEVICE)
    methods = {}
    for label, suffix, ttl, matrix_path in [
        ("InfoMatch", "_hybrid", "grafo_ttl_hibrido.ttl", "embed_matrix_hybrid.npy"),
        ("LinkGraph", "_hybrid_graphsage", "grafo_ttl_hibrido_graphsage.ttl", "embed_matrix_hybrid_graphsage.npy")
    ]:
        with open(f"id_map{suffix}.pkl", "rb") as f:
            id_map = pickle.load(f)
        with open(f"grafo_embed{suffix}.pickle", "rb") as f:
            G = pickle.load(f)
        matrix = np.load(matrix_path)
        rdf = RDFGraph()
        rdf.parse(ttl, format="ttl")
        methods[label] = (matrix, id_map, G, rdf)
    return methods, embedder

def get_top_k(matrix, id_map, query, k, embedder):
    vec = embedder.encode(f"query: {query}", convert_to_tensor=True, device=DEVICE)
    vec = vec.cpu().numpy().astype("float32")
    sims = np.dot(matrix, vec) / (np.linalg.norm(matrix, axis=1) * np.linalg.norm(vec) + 1e-10)
    top_k_idx = np.argsort(sims)[-k:][::-1]
    return [id_map[i] for i in top_k_idx]

def get_context(G, lang_id):
    node = G.nodes.get(lang_id, {})
    lines = [f"**Language:** {node.get('label', lang_id)}"]
    if node.get("wikipedia_summary"):
        lines.append(f"**Wikipedia:** {node['wikipedia_summary']}")
    if node.get("wikidata_description"):
        lines.append(f"**Wikidata:** {node['wikidata_description']}")
    if node.get("wikidata_countries"):
        lines.append(f"**Countries:** {node['wikidata_countries']}")
    return "\n\n".join(lines)

def query_rdf(rdf, lang_id):
    q = f"""
    PREFIX ex: <http://example.org/lang/>
    SELECT ?property ?value WHERE {{ ex:{lang_id} ?property ?value }}
    """
    try:
        return [(str(row[0]).split("/")[-1], str(row[1])) for row in rdf.query(q)]
    except Exception as e:
        return [("error", str(e))]

def generate_response(matrix, id_map, G, rdf, user_question, k, embedder):
    ids = get_top_k(matrix, id_map, user_question, k, embedder)
    context = [get_context(G, i) for i in ids]
    rdf_facts = []
    for i in ids:
        rdf_facts.extend([f"{p}: {v}" for p, v in query_rdf(rdf, i)])
    prompt = f"""<s>[INST]
You are an expert in South American indigenous languages.
Use strictly and only the information below to answer the user question in **English**.
- Do not infer or assume facts that are not explicitly stated.
- If the answer is unknown or insufficient, say \"I cannot answer with the available data.\"
- Limit your answer to 100 words.
### CONTEXT:
{chr(10).join(context)}
### RDF RELATIONS:
{chr(10).join(rdf_facts)}
### QUESTION:
{user_question}
Answer:
[/INST]"""
    try:
        res = requests.post(
            ENDPOINT_URL,
            headers={"Authorization": f"Bearer {HF_API_TOKEN}", "Content-Type": "application/json"},
            json={"inputs": prompt}, timeout=60
        )
        out = res.json()
        if isinstance(out, list) and "generated_text" in out[0]:
            return out[0]["generated_text"].replace(prompt.strip(), "").strip(), ids, context, rdf_facts
        return str(out), ids, context, rdf_facts
    except Exception as e:
        return str(e), ids, context, rdf_facts

# === MAIN APP ===
def main():
    methods, embedder = load_all_components()

    st.markdown("""
    <div class="header">
        <h1>๐ŸŒ Language Atlas: South American Indigenous Languages</h1>
    </div>
    """, unsafe_allow_html=True)
    
    with st.expander("๐Ÿ“Œ **Overview**", expanded=True):
        st.markdown("""
        This app provides **AI-powered analysis** of endangered indigenous languages in South America, 
        integrating knowledge graphs from **Glottolog, Wikipedia, and Wikidata**.
        \n\n*This is version 1 and currently English-only. Spanish version coming soon!*
        """)
    
    with st.sidebar:
        st.markdown("### ๐Ÿ“š Pontificia Universidad Catรณlica del Perรบ")
        st.markdown("""
        - <span class="tech-badge">Departamento de Humanidades</span>  
        - <span class="tech-badge">jveraz@pucp.edu.pe</span> 
        - <span class="tech-badge">Suggestions? Contact us</span>
        """, unsafe_allow_html=True)
        st.markdown("---")
        st.markdown("### ๐Ÿš€ Quick Start")
        st.markdown("""
        1. **Type a question** in the input box  
        2. **Click 'Analyze'** to compare methods  
        3. **Explore results** with expandable details  
        """)
        
        st.markdown("---")
        st.markdown("### ๐Ÿ” Example Queries")
        questions = [
            "What languages are endangered in Brazil?",
            "What languages are spoken in Perรบ?",
            "Which languages are related to Quechua?",
            "Where is Mapudungun spoken?"
        ]
        
        for q in questions:
            if st.markdown(f"<div class='suggested-question'>{q}</div>", unsafe_allow_html=True):
                st.session_state.query = q
        
        st.markdown("---")
        st.markdown("### โš™๏ธ Technical Details")
        st.markdown("""
        - <span class="tech-badge">Embeddings</span> Node2Vec vs. GraphSAGE  
        - <span class="tech-badge">Language Model</span> Mistral-7B-Instruct  
        - <span class="tech-badge">Knowledge Graph</span> RDF-based integration  
        """, unsafe_allow_html=True)
        
        st.markdown("---")
        st.markdown("### ๐Ÿ“‚ Data Sources")
        st.markdown("""
        - **Glottolog** (Language classification)  
        - **Wikipedia** (Textual summaries)  
        - **Wikidata** (Structured facts)  
        """)
        
        st.markdown("---")
        st.markdown("### ๐Ÿ“Š Analysis Parameters")
        k = st.slider("Number of languages to analyze", 1, 10, 3)
        st.markdown("---")
        st.markdown("### ๐Ÿ”ง Advanced Options")
        show_ctx = st.checkbox("Show context information", False)
        show_rdf = st.checkbox("Show structured facts", False)

    st.markdown("### ๐Ÿ“ Ask About Indigenous Languages")
    query = st.text_input(
        "Enter your question:",
        value=st.session_state.get("query", ""),
        label_visibility="collapsed",
        placeholder="e.g. What languages are spoken in Peru?"
    )

    if st.button("Analyze", type="primary", use_container_width=True):
        if not query:
            st.warning("Please enter a question")
            return
            
        col1, col2 = st.columns(2)
        
        for col, (label, method) in zip([col1, col2], methods.items()):
            with col:
                st.markdown(f"#### {label} Method")
                st.caption({
                    "InfoMatch": "Node2Vec embeddings combining text and graph structure",
                    "LinkGraph": "GraphSAGE embeddings capturing network patterns"
                }[label])
                
                start = datetime.datetime.now()
                response, lang_ids, context, rdf_data = generate_response(*method, query, k, embedder)
                duration = (datetime.datetime.now() - start).total_seconds()
                
                st.markdown(f"""
                <div class="response-card">
                    {response}
                    <div style="margin-top: 1rem;">
                        <span class="metric-badge">โฑ๏ธ {duration:.2f}s</span>
                        <span class="metric-badge">๐ŸŒ {len(lang_ids)} languages</span>
                    </div>
                </div>
                """, unsafe_allow_html=True)
                
                if show_ctx:
                    with st.expander(f"๐Ÿ“– Context from {len(lang_ids)} languages"):
                        for lang_id, ctx in zip(lang_ids, context):
                            st.markdown(f"<div class='language-card'>{ctx}</div>", unsafe_allow_html=True)
                
                if show_rdf:
                    with st.expander("๐Ÿ”— Structured facts (RDF)"):
                        st.code("\n".join(rdf_data))
        
        st.markdown("---")
        st.markdown("""
        <div style="font-size: 0.8rem; color: #64748b; text-align: center;">
        <b>๐Ÿ“Œ Note:</b> This tool is designed for researchers, linguists, and cultural preservationists.  
        For best results, use specific questions about languages, families, or regions.
        </div>
        """, unsafe_allow_html=True)

if __name__ == "__main__":
    main()