Venue Review & Event Experience Sentiment Monitor

Scrape TripAdvisor for event venues in any city worldwide, collect real visitor reviews, and get back structured sentiment intelligence: composite score, polarity breakdown, and thematic analysis across 10 experience categories. Supports 10 venue types from concert halls to wedding venues. No API key required.

How it works: The actor searches TripAdvisor for matching venues in each city, then visits each venue's review page to collect up to 50 real reviews. AFINN word-level sentiment scoring and thematic keyword extraction run locally — no external API needed. The composite score blends text-based AFINN analysis (40%) with TripAdvisor's star rating (60%) for a balanced signal.

---

What This Actor Does

Given a list of cities and a venue type, the actor:

1. Searches TripAdvisor for matching venues in each city
2. Visits each venue's detail page to collect real visitor reviews
3. Fetches additional review pages if maxReviewsPerVenue exceeds 10
4. Applies AFINN word-level sentiment scoring across all collected review text
5. Extracts thematic signals across 10 experience categories
6. Returns a ranked, structured record per venue — ready for comparison, export, or analysis

---

Supported Venue Types

Value	Label
live_music	Concert Hall & Live Music Venue
conference_centre	Conference Centre & Corporate Events
wedding_venue	Wedding Venue
private_event_space	Private Party & Event Space
theatre	Theatre & Performing Arts
nightclub	Nightclub & Club Venue
rooftop_venue	Rooftop Venue
outdoor_venue	Outdoor & Festival Venue
exhibition_hall	Exhibition Hall & Trade Show Space
sports_arena	Sports Arena & Stadium

---

Sentiment Analysis — Three Approaches, Zero API Costs

⭐ Composite Score

Blends AFINN text analysis (40%) with TripAdvisor's official star rating (60%) for a robust, balanced sentiment signal. When no star rating is available, falls back to AFINN-only scoring.

🧠 AFINN Word Scoring

Scores every review word against the AFINN-165 lexicon (3,382 rated English words). The aggregate is normalised to a 0–100 scale. Runs entirely locally — no external API call.

🎯 Thematic Extraction

Detects mentions of 10 venue experience themes across all collected reviews and classifies each theme as positive, neutral, or negative.

Themes detected:

Theme	What it covers
acoustics	Sound quality, echo, reverb, speaker system
staff	Service, security, hosts, professionalism
parking	Car parks, transport links, accessibility
catering	Food, drinks, bar quality, menus
capacity	Crowd density, seating, spaciousness
av_technical	Screens, lighting, stage, production
price	Ticket costs, drink prices, value for money
location	Proximity to city centre, transport convenience
atmosphere	Vibe, energy, ambiance, overall experience feel
cleanliness	Toilets, hygiene, maintenance

---

Example Output

{
  "city": "Lisbon",
  "venueType": "Concert Hall & Live Music Venue",
  "rank": 1,
  "name": "MEO Arena",
  "tripAdvisorUrl": "https://www.tripadvisor.com/Attraction_Review-g189158-d1234567-Reviews-MEO_Arena-Lisbon.html",
  "rating": 4.5,
  "reviewCount": 1240,
  "reviewsAnalysed": 10,
  "sentimentScore": 76,
  "sentimentLabel": "Very Positive",
  "positivePercentage": 80,
  "neutralPercentage": 10,
  "negativePercentage": 10,
  "themes": {
    "atmosphere":  { "mentions": 7, "sentiment": "positive" },
    "acoustics":   { "mentions": 5, "sentiment": "positive" },
    "location":    { "mentions": 4, "sentiment": "positive" },
    "catering":    { "mentions": 3, "sentiment": "neutral" },
    "parking":     { "mentions": 3, "sentiment": "negative" }
  },
  "source": "tripadvisor",
  "scrapedAt": "2026-02-25T10:00:00.000Z"
}

---

Sentiment Score Scale

Score	Label
70–100	Very Positive
55–69	Positive
40–54	Mixed
25–39	Negative
0–24	Very Negative

---

Use Cases

🎤 Event & Tour Planning

Identify the top venues in any city before booking a show, tour stop, or live event. Compare sentiment scores and thematic breakdowns side by side — see which venues consistently receive praise for acoustics or atmosphere, and which have recurring complaints about parking or pricing.

🏢 Corporate & Private Event Selection

Shortlist conference centres, private event spaces, and wedding venues across multiple cities worldwide. Thematic extraction surfaces what matters most to attendees — catering quality, AV reliability, accessibility — directly from real visitor reviews.

🔬 Market Research & Venue Benchmarking

Understand the competitive landscape for a venue type in any city. Run the actor across multiple cities to compare sentiment distributions globally and identify markets where venue quality consistently under- or over-delivers.

📊 Venue Management & Reputation Monitoring

Schedule the actor to run periodically on your own venue to track sentiment score changes over time. Early detection of theme-level shifts — e.g. rising negative mentions of cleanliness or staff — enables faster operational response before issues compound.

---

Input

Field	Type	Required	Description
cities	Array	✅	Cities to search — any city with TripAdvisor coverage (e.g. ["New York", "London", "Lisbon", "Tokyo"])
venueType	Enum	✅	Venue type from the dropdown list
topN	Number	❌	Useful venue rows to return per city. The actor crawls extra candidates to replace blocked or zero-review pages when possible (default: 10, max: 50)
maxReviewsPerVenue	Number	❌	Reviews to collect per venue for sentiment analysis (default: 10, max: 50)
includePartialResults	Boolean	❌	Include venues with metadata but zero collected reviews. Disabled by default for clean sentiment datasets
proxyConfiguration	Object	❌	Apify Residential proxy is used by default on the Apify platform to reduce TripAdvisor 403 blocks

API Example

curl -X POST "https://api.apify.com/v2/acts/trovevault~venue-review-event-experience-sentiment-monitor/runs" \
  -H "Authorization: Bearer $APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"cities":["London"],"venueType":"live_music","topN":3,"maxReviewsPerVenue":10}'

---

Output Fields

Field	Type	Description
city	String	City searched
venueType	String	Venue type label
rank	Number	Rank within the emitted city results after blocked or zero-review candidates are filtered
name	String	Venue name
tripAdvisorUrl	String	Direct link to the venue's TripAdvisor review page
rating	Number	TripAdvisor overall star rating (1–5)
reviewCount	Number	Total number of reviews on TripAdvisor
reviewsAnalysed	Number	Number of reviews collected and used for sentiment analysis
sentimentScore	Number	Composite sentiment score (0–100)
sentimentLabel	String	Very Positive / Positive / Mixed / Negative / Very Negative
positivePercentage	Number	% of analysed reviews with positive AFINN score
neutralPercentage	Number	% of analysed reviews with neutral AFINN score
negativePercentage	Number	% of analysed reviews with negative AFINN score
themes	Object	Per-theme mention count and sentiment classification
source	String	tripadvisor
partialResult	Boolean	Whether the row was emitted with zero collected reviews
scrapedAt	String	ISO 8601 timestamp

---

How It Works

1. PlaywrightCrawler opens a full Chromium browser for each request — bypassing Cloudflare JS challenges that block HTTP-only scrapers
2. Venue discovery searches TripAdvisor for the venue type + city query and extracts more Attraction_Review candidates than the requested topN
3. Review collection visits each venue's TripAdvisor page and collects up to maxReviewsPerVenue real visitor reviews
4. Pagination automatically fetches additional review pages when maxReviewsPerVenue > 10
5. AFINN scoring runs across all collected review text — word by word, entirely local
6. Thematic extraction scans reviews for 10 experience theme keywords and computes per-theme sentiment
7. Composite scoring combines AFINN normalised score (40%) and star rating positivity (60%)
8. Results with collected reviews are re-ranked and pushed to the Apify dataset until the requested topN is reached, when enough valid candidates are available

Proxy note: TripAdvisor often returns 403 blocks from cloud data-center IPs. On the Apify platform, this actor uses Apify Residential proxy by default when no proxy configuration is supplied. You can override this in the input proxy settings.

---

Changelog

v0.1 – Initial release. Worldwide coverage via TripAdvisor. PlaywrightCrawler for reliable anti-bot bypass. Supports 10 venue types, multi-city input, configurable review depth (10–50 per venue), AFINN sentiment scoring, and thematic extraction across 10 experience categories.