Spaces:
Sleeping
Sleeping
File size: 46,652 Bytes
66b9109 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 | """
Enhanced visualization components for the advanced analytics dashboard.
Creates interactive charts, plots, network graphs, Sankey diagrams, and visual analytics.
"""
import pandas as pd
import numpy as np
import plotly.express as px
import plotly.graph_objects as go
from plotly.subplots import make_subplots
import seaborn as sns
import matplotlib.pyplot as plt
from wordcloud import WordCloud
import streamlit as st
from typing import Dict, List, Any, Tuple, Optional
from datetime import datetime, timedelta
import base64
from io import BytesIO
import networkx as nx
from collections import defaultdict
class KPIEngine:
"""Handles KPI calculations and display for the dashboard header."""
@staticmethod
def calculate_kpis(df: pd.DataFrame, areas_of_improvement: pd.DataFrame,
strength_anchors: pd.DataFrame) -> Dict[str, Any]:
"""
Calculate all KPIs for the dashboard header.
Args:
df: Processed dataframe
areas_of_improvement: Problem areas dataframe
strength_anchors: Strength areas dataframe
Returns:
Dictionary with all KPI values
"""
total_reviews = len(df)
# Sentiment percentages
positive_count = (df['overall_sentiment'] == 'Positive').sum()
negative_count = (df['overall_sentiment'] == 'Negative').sum()
neutral_count = (df['overall_sentiment'] == 'Neutral').sum()
positive_pct = (positive_count / total_reviews * 100) if total_reviews > 0 else 0
negative_pct = (negative_count / total_reviews * 100) if total_reviews > 0 else 0
neutral_pct = (neutral_count / total_reviews * 100) if total_reviews > 0 else 0
# Problem areas and strength anchors counts
problem_areas_count = len(areas_of_improvement)
strength_anchors_count = len(strength_anchors)
# Language distribution
languages = df['detected_language'].value_counts().to_dict()
# Intent distribution
intent_distribution = df['intent'].value_counts().to_dict()
complaint_pct = (df['intent'] == 'complaint').mean() * 100
return {
'total_reviews': total_reviews,
'positive_pct': round(positive_pct, 1),
'negative_pct': round(negative_pct, 1),
'neutral_pct': round(neutral_pct, 1),
'positive_count': positive_count,
'negative_count': negative_count,
'neutral_count': neutral_count,
'problem_areas_count': problem_areas_count,
'strength_anchors_count': strength_anchors_count,
'languages': languages,
'intent_distribution': intent_distribution,
'complaint_pct': round(complaint_pct, 1)
}
@staticmethod
def create_kpi_header(kpis: Dict[str, Any]) -> None:
"""
Create and display the KPI header section.
Args:
kpis: Dictionary containing all KPI values
"""
# Main KPI Row
col1, col2, col3, col4, col5 = st.columns(5)
with col1:
st.metric(
label="π Total Reviews",
value=f"{kpis['total_reviews']:,}",
delta=None
)
with col2:
st.metric(
label="π Positive %",
value=f"{kpis['positive_pct']}%",
delta=f"{kpis['positive_count']} reviews"
)
with col3:
st.metric(
label="π Negative %",
value=f"{kpis['negative_pct']}%",
delta=f"{kpis['negative_count']} reviews"
)
with col4:
st.metric(
label="π΄ Problem Areas",
value=kpis['problem_areas_count'],
delta="Aspects needing attention"
)
with col5:
st.metric(
label="π’ Strength Anchors",
value=kpis['strength_anchors_count'],
delta="Positive aspects"
)
class AdvancedVisualizationEngine:
"""Enhanced visualization engine with advanced charts and network analysis."""
def __init__(self):
self.color_schemes = {
'sentiment': {'Positive': '#2E8B57', 'Negative': '#DC143C', 'Neutral': '#708090'},
'intent': px.colors.qualitative.Set3,
'default': px.colors.qualitative.Plotly,
'priority': px.colors.sequential.Reds,
'strength': px.colors.sequential.Greens
}
def create_dual_ranking_tables(self, areas_of_improvement: pd.DataFrame,
strength_anchors: pd.DataFrame) -> None:
"""
Create and display the dual ranking tables.
Args:
areas_of_improvement: Problem areas dataframe
strength_anchors: Strength areas dataframe
"""
col1, col2 = st.columns(2)
with col1:
st.subheader("π΄ Areas of Improvement")
if len(areas_of_improvement) > 0:
# Format the dataframe for display
display_improvement = areas_of_improvement.copy()
display_improvement['Negativity %'] = display_improvement['negativity_pct'].astype(str) + '%'
display_improvement['Intent Severity'] = display_improvement['intent_severity']
display_improvement['Frequency'] = display_improvement['frequency']
display_improvement['Priority Score'] = display_improvement['priority_score']
# Select columns for display
display_cols = ['aspect', 'Negativity %', 'Intent Severity', 'Frequency', 'Priority Score']
display_improvement = display_improvement[display_cols].rename(columns={'aspect': 'Aspect'})
st.dataframe(
display_improvement,
use_container_width=True,
hide_index=True
)
else:
st.info("No significant problem areas identified.")
with col2:
st.subheader("π’ Strength Anchors")
if len(strength_anchors) > 0:
# Format the dataframe for display
display_strength = strength_anchors.copy()
display_strength['Positivity %'] = display_strength['positivity_pct'].astype(str) + '%'
display_strength['Intent Type'] = display_strength['intent_type']
display_strength['Frequency'] = display_strength['frequency']
display_strength['Strength Score'] = display_strength['strength_score']
# Select columns for display
display_cols = ['aspect', 'Positivity %', 'Intent Type', 'Frequency', 'Strength Score']
display_strength = display_strength[display_cols].rename(columns={'aspect': 'Aspect'})
st.dataframe(
display_strength,
use_container_width=True,
hide_index=True
)
else:
st.info("No significant strength anchors identified.")
def create_aspect_network_graph(self, aspect_network: nx.Graph) -> go.Figure:
"""
Create interactive network graph showing aspect relationships.
Args:
aspect_network: NetworkX graph with aspect co-occurrence data
Returns:
Plotly figure object
"""
if len(aspect_network.nodes()) == 0:
fig = go.Figure()
fig.add_annotation(
text="No aspect relationships found",
xref="paper", yref="paper",
x=0.5, y=0.5, xanchor='center', yanchor='middle',
showarrow=False, font_size=16
)
fig.update_layout(title="Aspect Co-occurrence Network")
return fig
# Calculate positions using spring layout
pos = nx.spring_layout(aspect_network, k=3, iterations=50)
# Extract node information
node_x = []
node_y = []
node_text = []
node_size = []
node_color = []
node_info = []
for node in aspect_network.nodes():
x, y = pos[node]
node_x.append(x)
node_y.append(y)
# Get node attributes
freq = aspect_network.nodes[node].get('frequency', 1)
sentiment_score = aspect_network.nodes[node].get('sentiment_score', 0)
color = aspect_network.nodes[node].get('color', 'gray')
positive_pct = aspect_network.nodes[node].get('positive_pct', 0) * 100
negative_pct = aspect_network.nodes[node].get('negative_pct', 0) * 100
node_text.append(node)
node_size.append(max(20, freq * 5)) # Size based on frequency
# Color based on sentiment
if color == 'green':
node_color.append('#2E8B57')
elif color == 'red':
node_color.append('#DC143C')
else:
node_color.append('#708090')
node_info.append(f"Aspect: {node}<br>" +
f"Frequency: {freq}<br>" +
f"Positive: {positive_pct:.1f}%<br>" +
f"Negative: {negative_pct:.1f}%")
# Extract edge information
edge_x = []
edge_y = []
edge_weights = []
for edge in aspect_network.edges():
x0, y0 = pos[edge[0]]
x1, y1 = pos[edge[1]]
edge_x.extend([x0, x1, None])
edge_y.extend([y0, y1, None])
weight = aspect_network.edges[edge].get('weight', 1)
edge_weights.append(weight)
# Create edge trace
edge_trace = go.Scatter(
x=edge_x, y=edge_y,
line=dict(width=2, color='#888'),
hoverinfo='none',
mode='lines'
)
# Create node trace
node_trace = go.Scatter(
x=node_x, y=node_y,
mode='markers+text',
hoverinfo='text',
text=node_text,
hovertext=node_info,
textposition="middle center",
marker=dict(
size=node_size,
color=node_color,
line=dict(width=2, color='white')
)
)
# Create figure
fig = go.Figure(data=[edge_trace, node_trace],
layout=go.Layout(
title=dict(text="Aspect Co-occurrence Network", font=dict(size=16)),
showlegend=False,
hovermode='closest',
margin=dict(b=20,l=5,r=5,t=40),
annotations=[ dict(
text="Node size = frequency, Color = sentiment (green=positive, red=negative)",
showarrow=False,
xref="paper", yref="paper",
x=0.005, y=-0.002,
xanchor='left', yanchor='bottom',
font=dict(color='gray', size=10)
)],
xaxis=dict(showgrid=False, zeroline=False, showticklabels=False),
yaxis=dict(showgrid=False, zeroline=False, showticklabels=False),
height=600
))
return fig
def create_intent_aspect_sankey(self, df: pd.DataFrame) -> go.Figure:
"""
Create Sankey diagram showing Intent β Aspect β Sentiment flow.
Args:
df: Processed dataframe
Returns:
Plotly figure object
"""
# Prepare data for Sankey diagram
sankey_data = []
for idx, row in df.iterrows():
aspects = row['aspects'] if isinstance(row['aspects'], list) else []
sentiments = row['aspect_sentiments'] if isinstance(row['aspect_sentiments'], list) else []
intent = row['intent']
for aspect, sentiment in zip(aspects, sentiments):
sankey_data.append({
'intent': intent,
'aspect': aspect,
'sentiment': sentiment
})
if not sankey_data:
fig = go.Figure()
fig.add_annotation(
text="No data available for Sankey diagram",
xref="paper", yref="paper",
x=0.5, y=0.5, xanchor='center', yanchor='middle',
showarrow=False, font_size=16
)
fig.update_layout(title="Intent β Aspect β Sentiment Flow")
return fig
sankey_df = pd.DataFrame(sankey_data)
# Create node lists
intents = sorted(sankey_df['intent'].unique())
aspects = sorted(sankey_df['aspect'].unique())
sentiments = sorted(sankey_df['sentiment'].unique())
# Create labels and indices
all_labels = intents + aspects + sentiments
label_to_idx = {label: idx for idx, label in enumerate(all_labels)}
# Create flows
flows = []
# Intent β Aspect flows
intent_aspect_flows = sankey_df.groupby(['intent', 'aspect']).size().reset_index(name='count')
for _, row in intent_aspect_flows.iterrows():
flows.append({
'source': label_to_idx[row['intent']],
'target': label_to_idx[row['aspect']],
'value': row['count']
})
# Aspect β Sentiment flows
aspect_sentiment_flows = sankey_df.groupby(['aspect', 'sentiment']).size().reset_index(name='count')
for _, row in aspect_sentiment_flows.iterrows():
flows.append({
'source': label_to_idx[row['aspect']],
'target': label_to_idx[row['sentiment']],
'value': row['count']
})
# Create colors
intent_colors = ['rgba(255, 127, 14, 0.8)'] * len(intents)
aspect_colors = ['rgba(31, 119, 180, 0.8)'] * len(aspects)
sentiment_colors = []
for sentiment in sentiments:
if sentiment == 'Positive':
sentiment_colors.append('rgba(46, 139, 87, 0.8)')
elif sentiment == 'Negative':
sentiment_colors.append('rgba(220, 20, 60, 0.8)')
else:
sentiment_colors.append('rgba(112, 128, 144, 0.8)')
node_colors = intent_colors + aspect_colors + sentiment_colors
# Create Sankey diagram
fig = go.Figure(data=[go.Sankey(
node=dict(
pad=15,
thickness=20,
line=dict(color="black", width=0.5),
label=all_labels,
color=node_colors
),
link=dict(
source=[flow['source'] for flow in flows],
target=[flow['target'] for flow in flows],
value=[flow['value'] for flow in flows]
)
)])
fig.update_layout(
title=dict(text="Intent β Aspect β Sentiment Flow", font=dict(size=12)),
font_size=12,
height=600
)
return fig
def create_enhanced_timeline_chart(self, df: pd.DataFrame,
annotations: Optional[List[Dict]] = None) -> go.Figure:
"""
Create enhanced timeline chart with annotation support.
Args:
df: Processed dataframe with date and sentiment columns
annotations: List of event annotations to add to chart
Returns:
Plotly figure object
"""
try:
# Ensure we have data
if df.empty:
fig = go.Figure()
fig.add_annotation(
text="No data available for timeline chart",
xref="paper", yref="paper",
x=0.5, y=0.5, xanchor='center', yanchor='middle',
showarrow=False, font_size=16
)
fig.update_layout(title="Sentiment Trends Over Time")
return fig
# Ensure date column is properly formatted
df_copy = df.copy()
df_copy['date'] = pd.to_datetime(df_copy['date'], errors='coerce')
# Remove any rows with invalid dates
df_copy = df_copy.dropna(subset=['date'])
if df_copy.empty:
fig = go.Figure()
fig.add_annotation(
text="No valid dates found in data",
xref="paper", yref="paper",
x=0.5, y=0.5, xanchor='center', yanchor='middle',
showarrow=False, font_size=16
)
fig.update_layout(title="Sentiment Trends Over Time")
return fig
# Group by date and sentiment
timeline_data = df_copy.groupby([df_copy['date'].dt.date, 'overall_sentiment']).size().reset_index(name='count')
timeline_data['date'] = pd.to_datetime(timeline_data['date'])
# Sort by date for proper line chart
timeline_data = timeline_data.sort_values('date')
# Create figure manually
fig = go.Figure()
# Add traces for each sentiment
sentiments = timeline_data['overall_sentiment'].unique()
for sentiment in sentiments:
sentiment_data = timeline_data[timeline_data['overall_sentiment'] == sentiment].copy()
fig.add_trace(go.Scatter(
x=sentiment_data['date'],
y=sentiment_data['count'],
mode='lines+markers',
name=sentiment,
line=dict(
color=self.color_schemes['sentiment'].get(sentiment, '#999999'),
width=2
),
marker=dict(size=6),
hovertemplate=f'<b>{sentiment}</b><br>Date: %{{x|%Y-%m-%d}}<br>Count: %{{y}}<extra></extra>'
))
# Add simple text annotations (no arrows or complex shapes)
if annotations and len(timeline_data) > 0:
max_count = timeline_data['count'].max()
y_range = max_count * 0.3 # Space for annotations
for i, annotation in enumerate(annotations):
try:
# Convert annotation date
if isinstance(annotation['date'], str):
ann_date = pd.to_datetime(annotation['date'])
else:
ann_date = pd.to_datetime(annotation['date'])
# Add simple text annotation
fig.add_annotation(
x=ann_date,
y=max_count + y_range * (0.2 + i * 0.3),
text=f"π {annotation['text']}",
showarrow=False,
bgcolor="rgba(255,255,255,0.9)",
bordercolor="purple",
borderwidth=1,
font=dict(size=10, color="purple"),
xanchor="center"
)
except Exception:
# Skip problematic annotations silently
continue
fig.update_layout(
title="Sentiment Trends Over Time",
xaxis_title="Date",
yaxis_title="Number of Reviews",
template='plotly_white',
height=500,
showlegend=True,
xaxis=dict(
tickangle=45,
type='date'
),
yaxis=dict(
rangemode='tozero'
),
hovermode='x unified'
)
return fig
except Exception as e:
# Fallback: return empty chart with error message
fig = go.Figure()
fig.add_annotation(
text=f"Error creating timeline chart: {str(e)}",
xref="paper", yref="paper",
x=0.5, y=0.5, xanchor='center', yanchor='middle',
showarrow=False, font_size=14, font_color="red"
)
fig.update_layout(title="Sentiment Trends Over Time")
return fig
def create_regional_language_analysis(self, df: pd.DataFrame) -> go.Figure:
"""
Create comprehensive language-wise sentiment and intent analysis.
Args:
df: Processed dataframe
Returns:
Plotly figure object with subplots
"""
# Create subplots
fig = make_subplots(
rows=2, cols=2,
subplot_titles=('Language Distribution', 'Sentiment by Language',
'Intent by Language', 'Language Trends'),
specs=[[{"type": "pie"}, {"type": "bar"}],
[{"type": "bar"}, {"type": "scatter"}]]
)
# 1. Language Distribution (Pie Chart)
lang_counts = df['detected_language'].value_counts()
fig.add_trace(
go.Pie(labels=lang_counts.index, values=lang_counts.values, name="Languages"),
row=1, col=1
)
# 2. Sentiment by Language (Stacked Bar)
sentiment_lang = df.groupby(['detected_language', 'overall_sentiment']).size().unstack(fill_value=0)
for sentiment in sentiment_lang.columns:
fig.add_trace(
go.Bar(
name=sentiment,
x=sentiment_lang.index,
y=sentiment_lang[sentiment],
marker_color=self.color_schemes['sentiment'].get(sentiment, '#808080')
),
row=1, col=2
)
# 3. Intent by Language (Stacked Bar)
intent_lang = df.groupby(['detected_language', 'intent']).size().unstack(fill_value=0)
for intent in intent_lang.columns:
fig.add_trace(
go.Bar(
name=intent,
x=intent_lang.index,
y=intent_lang[intent],
showlegend=False
),
row=2, col=1
)
# 4. Language Trends Over Time
df_copy = df.copy()
df_copy['date'] = pd.to_datetime(df_copy['date'])
df_copy['date_str'] = df_copy['date'].dt.strftime('%Y-%m-%d')
daily_lang = df_copy.groupby(['date_str', 'detected_language']).size().unstack(fill_value=0)
daily_lang.index = pd.to_datetime(daily_lang.index)
for lang in daily_lang.columns:
fig.add_trace(
go.Scatter(
x=daily_lang.index,
y=daily_lang[lang],
mode='lines',
name=f'{lang} trend',
showlegend=False
),
row=2, col=2
)
fig.update_layout(
height=800,
title=dict(text="Regional/Language Analysis Dashboard", font=dict(size=16)),
showlegend=True
)
return fig
def create_alert_dashboard(self, sentiment_alerts: List[Dict[str, Any]]) -> None:
"""
Create alert dashboard showing sentiment spikes.
Args:
sentiment_alerts: List of sentiment spike alerts
"""
st.subheader("π¨ Sentiment Alerts")
if not sentiment_alerts:
st.info("No sentiment spikes detected in recent data.")
return
# Display alerts
for i, alert in enumerate(sentiment_alerts[:5]): # Show top 5 alerts
severity_color = "π΄" if alert['alert_severity'] == 'high' else "π‘"
with st.expander(f"{severity_color} {alert['aspect']} - {alert['spike_magnitude']}% increase"):
col1, col2, col3 = st.columns(3)
with col1:
st.metric("Recent Avg Negative", f"{alert['recent_avg_negative']:.1f}")
with col2:
st.metric("Previous Avg Negative", f"{alert['previous_avg_negative']:.1f}")
with col3:
st.metric("Spike Magnitude", f"{alert['spike_magnitude']}%")
st.warning(f"Aspect '{alert['aspect']}' showing {alert['alert_severity']} severity spike in negative sentiment.")
def create_impact_simulation_tool(self, df: pd.DataFrame) -> None:
"""
Create what-if analysis tool for aspect improvements.
Args:
df: Processed dataframe
"""
st.subheader("π― Impact Simulation")
st.write("Simulate the impact of fixing specific aspects on overall sentiment.")
# Get list of negative aspects
negative_aspects = []
for idx, row in df.iterrows():
aspects = row['aspects'] if isinstance(row['aspects'], list) else []
sentiments = row['aspect_sentiments'] if isinstance(row['aspect_sentiments'], list) else []
for aspect, sentiment in zip(aspects, sentiments):
if sentiment == 'Negative':
negative_aspects.append(aspect)
if not negative_aspects:
st.info("No negative aspects found for simulation.")
return
unique_negative_aspects = list(set(negative_aspects))
# Aspect selection
selected_aspects = st.multiselect(
"Select aspects to 'fix' (simulate removing negative reviews):",
unique_negative_aspects,
default=unique_negative_aspects[:3] if len(unique_negative_aspects) >= 3 else unique_negative_aspects
)
if selected_aspects:
# Calculate current sentiment distribution
current_sentiment = df['overall_sentiment'].value_counts()
current_positive_pct = (current_sentiment.get('Positive', 0) / len(df)) * 100
current_negative_pct = (current_sentiment.get('Negative', 0) / len(df)) * 100
# Simulate fixing aspects (remove reviews with selected negative aspects)
df_simulated = df.copy()
for aspect in selected_aspects:
# Remove reviews that mention this aspect negatively
mask = df_simulated.apply(lambda row: not (
isinstance(row['aspects'], list) and
isinstance(row['aspect_sentiments'], list) and
any(asp == aspect and sent == 'Negative'
for asp, sent in zip(row['aspects'], row['aspect_sentiments']))
), axis=1)
df_simulated = df_simulated[mask]
# Calculate new sentiment distribution
if len(df_simulated) > 0:
new_sentiment = df_simulated['overall_sentiment'].value_counts()
new_positive_pct = (new_sentiment.get('Positive', 0) / len(df_simulated)) * 100
new_negative_pct = (new_sentiment.get('Negative', 0) / len(df_simulated)) * 100
# Display results
col1, col2 = st.columns(2)
with col1:
st.metric("Current Positive %", f"{current_positive_pct:.1f}%")
st.metric("Current Negative %", f"{current_negative_pct:.1f}%")
with col2:
positive_change = new_positive_pct - current_positive_pct
negative_change = new_negative_pct - current_negative_pct
st.metric(
"Simulated Positive %",
f"{new_positive_pct:.1f}%",
delta=f"{positive_change:+.1f}%"
)
st.metric(
"Simulated Negative %",
f"{new_negative_pct:.1f}%",
delta=f"{negative_change:+.1f}%"
)
reviews_removed = len(df) - len(df_simulated)
st.info(f"Simulation removed {reviews_removed} negative reviews mentioning the selected aspects.")
def create_summary_sections(self, macro_summary: Dict[str, str],
micro_summaries: Dict[str, str]) -> None:
"""
Create and display macro and micro summary sections.
Args:
macro_summary: High-level insights
micro_summaries: Aspect-specific insights
"""
# Macro Summary
st.subheader("π Executive Summary")
for category, summary in macro_summary.items():
st.write(f"**{category.replace('_', ' ').title()}:** {summary}")
st.divider()
# Micro Summaries
st.subheader("π Aspect-Level Insights")
if micro_summaries:
for aspect, summary in micro_summaries.items():
with st.expander(f"π {aspect.title()} Analysis"):
st.write(summary)
else:
st.info("No detailed aspect summaries available.")
class ExportEngine:
"""Handles export functionality for reports and insights."""
@staticmethod
def generate_pdf_report(df: pd.DataFrame, kpis: Dict[str, Any],
areas_of_improvement: pd.DataFrame,
strength_anchors: pd.DataFrame) -> bytes:
"""
Generate PDF report with key insights.
Args:
df: Processed dataframe
kpis: KPI dictionary
areas_of_improvement: Problem areas dataframe
strength_anchors: Strength areas dataframe
Returns:
PDF bytes
"""
# Placeholder for PDF generation
# In a real implementation, you would use libraries like reportlab or weasyprint
pdf_content = f"""
SENTIMENT ANALYSIS REPORT
========================
Executive Summary:
- Total Reviews: {kpis['total_reviews']}
- Positive Sentiment: {kpis['positive_pct']}%
- Negative Sentiment: {kpis['negative_pct']}%
- Problem Areas: {kpis['problem_areas_count']}
- Strength Anchors: {kpis['strength_anchors_count']}
Top Issues:
{areas_of_improvement[['aspect', 'priority_score']].to_string() if len(areas_of_improvement) > 0 else 'None'}
Top Strengths:
{strength_anchors[['aspect', 'strength_score']].to_string() if len(strength_anchors) > 0 else 'None'}
"""
return pdf_content.encode('utf-8')
@staticmethod
def generate_excel_export(df: pd.DataFrame, areas_of_improvement: pd.DataFrame,
strength_anchors: pd.DataFrame) -> bytes:
"""
Generate Excel export with multiple sheets.
Args:
df: Processed dataframe
areas_of_improvement: Problem areas dataframe
strength_anchors: Strength areas dataframe
Returns:
Excel bytes
"""
from io import BytesIO
import pandas as pd
output = BytesIO()
with pd.ExcelWriter(output, engine='openpyxl') as writer:
df.to_excel(writer, sheet_name='Raw Data', index=False)
areas_of_improvement.to_excel(writer, sheet_name='Problem Areas', index=False)
strength_anchors.to_excel(writer, sheet_name='Strengths', index=False)
return output.getvalue()
def create_timeline_chart(self, df: pd.DataFrame) -> go.Figure:
"""
Create timeline chart showing sentiment trends over time.
Args:
df: Processed dataframe with date and sentiment columns
Returns:
Plotly figure object
"""
# Ensure date column is properly formatted
df_copy = df.copy()
df_copy['date'] = pd.to_datetime(df_copy['date'])
# Group by date and sentiment - convert to string dates to avoid timestamp issues
df_copy['date_str'] = df_copy['date'].dt.strftime('%Y-%m-%d')
timeline_data = df_copy.groupby(['date_str', 'overall_sentiment']).size().reset_index(name='count')
# Convert date strings back to datetime objects for plotting
timeline_data['date'] = pd.to_datetime(timeline_data['date_str'])
# Create the line chart
fig = go.Figure()
# Add traces for each sentiment
for sentiment in timeline_data['overall_sentiment'].unique():
sentiment_data = timeline_data[timeline_data['overall_sentiment'] == sentiment]
fig.add_trace(go.Scatter(
x=sentiment_data['date'],
y=sentiment_data['count'],
mode='lines+markers',
name=sentiment,
line=dict(color=self.color_schemes['sentiment'].get(sentiment, '#999999')),
marker=dict(size=6)
))
fig.update_layout(
title="Sentiment Trends Over Time",
xaxis_title="Date",
yaxis_title="Number of Reviews",
hovermode='x unified',
template='plotly_white',
height=400,
showlegend=True
)
return fig
def create_sentiment_distribution(self, df: pd.DataFrame) -> go.Figure:
"""Create pie chart for sentiment distribution."""
sentiment_counts = df['overall_sentiment'].value_counts()
fig = go.Figure(data=[go.Pie(
labels=sentiment_counts.index,
values=sentiment_counts.values,
hole=.3,
marker_colors=[self.color_schemes['sentiment'].get(label, '#808080') for label in sentiment_counts.index]
)])
fig.update_traces(textposition='inside', textinfo='percent+label')
fig.update_layout(
title="Overall Sentiment Distribution",
template='plotly_white'
)
return fig
def create_intent_distribution(self, df: pd.DataFrame) -> go.Figure:
"""Create bar chart for intent distribution."""
intent_counts = df['intent'].value_counts()
fig = px.bar(
x=intent_counts.index,
y=intent_counts.values,
title="Intent Classification Distribution",
labels={'x': 'Intent Type', 'y': 'Number of Reviews'},
color=intent_counts.index,
color_discrete_sequence=self.color_schemes['intent']
)
fig.update_layout(
showlegend=False,
template='plotly_white'
)
return fig
def create_aspect_sentiment_heatmap(self, df: pd.DataFrame) -> go.Figure:
"""
Create heatmap showing sentiment distribution across aspects.
Args:
df: Processed dataframe with aspects and aspect_sentiments
Returns:
Plotly figure object
"""
# Extract all aspects and their sentiments
aspect_sentiment_data = []
for idx, row in df.iterrows():
aspects = row['aspects'] if isinstance(row['aspects'], list) else []
sentiments = row['aspect_sentiments'] if isinstance(row['aspect_sentiments'], list) else []
for aspect, sentiment in zip(aspects, sentiments):
aspect_sentiment_data.append({'aspect': aspect, 'sentiment': sentiment})
if not aspect_sentiment_data:
# Return empty heatmap if no aspect data
fig = go.Figure()
fig.add_annotation(
text="No aspect data available for heatmap",
xref="paper", yref="paper",
x=0.5, y=0.5, xanchor='center', yanchor='middle',
showarrow=False, font_size=16
)
fig.update_layout(title="Aspect-Sentiment Heatmap")
return fig
# Create dataframe and pivot table
aspect_df = pd.DataFrame(aspect_sentiment_data)
heatmap_data = aspect_df.groupby(['aspect', 'sentiment']).size().unstack(fill_value=0)
# Create heatmap
fig = go.Figure(data=go.Heatmap(
z=heatmap_data.values,
x=heatmap_data.columns,
y=heatmap_data.index,
colorscale='RdYlGn',
text=heatmap_data.values,
texttemplate="%{text}",
textfont={"size": 12},
))
fig.update_layout(
title="Aspect-Sentiment Correlation Heatmap",
xaxis_title="Sentiment",
yaxis_title="Aspects",
template='plotly_white'
)
return fig
def create_wordcloud(self, df: pd.DataFrame, sentiment_filter: str = None) -> str:
"""
Create word cloud from reviews.
Args:
df: Processed dataframe
sentiment_filter: Filter by sentiment ('Positive', 'Negative', 'Neutral')
Returns:
Base64 encoded image string
"""
# Filter data if sentiment filter is provided
if sentiment_filter:
filtered_df = df[df['overall_sentiment'] == sentiment_filter]
else:
filtered_df = df
# Combine all translated reviews
text = ' '.join(filtered_df['translated_review'].astype(str))
if not text.strip():
return None
# Generate word cloud
wordcloud = WordCloud(
width=800,
height=400,
background_color='white',
max_words=100,
colormap='viridis'
).generate(text)
# Convert to base64 for display
img = BytesIO()
wordcloud.to_image().save(img, format='PNG')
img.seek(0)
img_b64 = base64.b64encode(img.read()).decode()
return img_b64
def create_language_distribution(self, df: pd.DataFrame) -> go.Figure:
"""Create pie chart for language distribution."""
lang_counts = df['detected_language'].value_counts()
fig = go.Figure(data=[go.Pie(
labels=lang_counts.index,
values=lang_counts.values,
hole=.3
)])
fig.update_traces(textposition='inside', textinfo='percent+label')
fig.update_layout(
title="Language Distribution",
template='plotly_white'
)
return fig
def create_correlation_matrix(self, df: pd.DataFrame) -> go.Figure:
"""
Create correlation matrix for numerical relationships.
Args:
df: Processed dataframe
Returns:
Plotly figure object
"""
# Create numerical features for correlation
correlation_df = pd.DataFrame()
# Add sentiment as numerical
sentiment_mapping = {'Positive': 1, 'Neutral': 0, 'Negative': -1}
correlation_df['sentiment_score'] = df['overall_sentiment'].map(sentiment_mapping)
# Add intent as categorical numerical
intent_mapping = {intent: idx for idx, intent in enumerate(df['intent'].unique())}
correlation_df['intent_code'] = df['intent'].map(intent_mapping)
# Add review length
correlation_df['review_length'] = df['translated_review'].str.len()
# Add number of aspects
correlation_df['aspect_count'] = df['aspects'].apply(lambda x: len(x) if isinstance(x, list) else 0)
# Calculate correlation matrix
corr_matrix = correlation_df.corr()
# Create heatmap
fig = go.Figure(data=go.Heatmap(
z=corr_matrix.values,
x=corr_matrix.columns,
y=corr_matrix.columns,
colorscale='RdBu',
zmid=0,
text=np.round(corr_matrix.values, 2),
texttemplate="%{text}",
textfont={"size": 12},
))
fig.update_layout(
title="Feature Correlation Matrix",
template='plotly_white'
)
return fig
def create_aspect_frequency_chart(self, df: pd.DataFrame) -> go.Figure:
"""Create bar chart showing most frequent aspects."""
# Extract all aspects
all_aspects = []
for aspects in df['aspects']:
if isinstance(aspects, list):
all_aspects.extend(aspects)
if not all_aspects:
fig = go.Figure()
fig.add_annotation(
text="No aspects extracted from reviews",
xref="paper", yref="paper",
x=0.5, y=0.5, xanchor='center', yanchor='middle',
showarrow=False, font_size=16
)
fig.update_layout(title="Most Frequent Aspects")
return fig
# Count frequency
aspect_counts = pd.Series(all_aspects).value_counts().head(15)
fig = px.bar(
x=aspect_counts.values,
y=aspect_counts.index,
orientation='h',
title="Most Frequent Aspects (Top 15)",
labels={'x': 'Frequency', 'y': 'Aspects'}
)
fig.update_layout(
template='plotly_white',
height=500
)
return fig
def create_daily_volume_chart(self, df: pd.DataFrame) -> go.Figure:
"""Create chart showing daily review volume."""
df_copy = df.copy()
df_copy['date'] = pd.to_datetime(df_copy['date'])
df_copy['date_str'] = df_copy['date'].dt.strftime('%Y-%m-%d')
daily_counts = df_copy.groupby('date_str').size().reset_index(name='count')
daily_counts['date'] = pd.to_datetime(daily_counts['date_str'])
daily_counts.columns = ['date_str', 'review_count', 'date']
fig = px.bar(
daily_counts,
x='date',
y='review_count',
title="Daily Review Volume",
labels={'review_count': 'Number of Reviews', 'date': 'Date'}
)
fig.update_layout(
template='plotly_white',
xaxis_title="Date",
yaxis_title="Number of Reviews"
)
return fig
class FilterEngine:
"""Handles data filtering based on user selections."""
@staticmethod
def apply_filters(df: pd.DataFrame, filters: Dict[str, Any]) -> pd.DataFrame:
"""
Apply multiple filters to the dataframe.
Args:
df: Original dataframe
filters: Dictionary containing filter criteria
Returns:
Filtered dataframe
"""
filtered_df = df.copy()
# Date range filter
if filters.get('date_range'):
start_date, end_date = filters['date_range']
# Convert to pandas datetime for comparison
filtered_df['date'] = pd.to_datetime(filtered_df['date'])
start_date_pd = pd.to_datetime(start_date)
end_date_pd = pd.to_datetime(end_date)
filtered_df = filtered_df[
(filtered_df['date'] >= start_date_pd) &
(filtered_df['date'] <= end_date_pd)
]
# Sentiment filter
if filters.get('sentiment') and filters['sentiment'] != 'All':
filtered_df = filtered_df[filtered_df['overall_sentiment'] == filters['sentiment']]
# Intent filter
if filters.get('intent') and filters['intent'] != 'All':
filtered_df = filtered_df[filtered_df['intent'] == filters['intent']]
# Language filter
if filters.get('language') and filters['language'] != 'All':
filtered_df = filtered_df[filtered_df['detected_language'] == filters['language']]
# Aspect filter
if filters.get('aspect') and filters['aspect'] != 'All':
filtered_df = filtered_df[
filtered_df['aspects'].apply(
lambda x: filters['aspect'] in x if isinstance(x, list) else False
)
]
return filtered_df
@staticmethod
def get_filter_options(df: pd.DataFrame) -> Dict[str, List]:
"""
Extract available filter options from the dataframe.
Args:
df: Processed dataframe
Returns:
Dictionary containing filter options
"""
# Extract unique aspects
all_aspects = set()
for aspects in df['aspects']:
if isinstance(aspects, list):
all_aspects.update(aspects)
return {
'sentiments': ['All'] + sorted(df['overall_sentiment'].unique().tolist()),
'intents': ['All'] + sorted(df['intent'].unique().tolist()),
'languages': ['All'] + sorted(df['detected_language'].unique().tolist()),
'aspects': ['All'] + sorted(list(all_aspects)),
'date_range': (df['date'].min().date(), df['date'].max().date())
} |