tracking-brand-sentiment-across-platforms …

원본 Skill 문서

원본 저장소의 제목, 예시, 코드, 표, 링크, 이미지를 유지해 표시합니다.

Tracking Brand Sentiment Across Platforms

Monitors brand sentiment on Twitter, Reddit, and TikTok in parallel, then produces a unified brand health score. Each platform serves a different role: Twitter = real-time news/opinion, Reddit = deep community discussion, TikTok = Gen Z product culture.

Prerequisites

APIFY_TOKEN environment variable set
Optional: Apify MCP server installed

Inputs

Parameter	Type	Required	Default	Notes
`searchTerms`	array	✅	`[]`	Twitter advanced search queries (e.g. `["#AI lang:en", "from:NASA"]`)
`sort`	string	Optional	`Top`	Sort order: `Latest`, `Top`, or `Latest+Top`
`tweetLanguage`	string	Optional	—	ISO 639-1 language code (e.g. `en`)
`maxItems`	number	Optional	Unlimited	Maximum tweets to return
`onlyVerifiedUsers`	boolean	Optional	`false`	Only tweets from verified users
`onlyTwitterBlue`	boolean	Optional	`false`	Only Twitter Blue subscribers
`onlyImage`	boolean	Optional	`false`	Only tweets with images
`onlyVideo`	boolean	Optional	`false`	Only tweets with videos
`onlyQuote`	boolean	Optional	`false`	Only quote tweets
`author`	string	Optional	—	Filter to a specific author handle
`inReplyTo`	string	Optional	—	Tweets replying to a specific handle
`mentioning`	string	Optional	—	Tweets mentioning a specific handle
`geotaggedNear`	string	Optional	—	Tweets near a location
`withinRadius`	string	Optional	—	Radius around geotaggedNear
`geocode`	string	Optional	—	Lat/lng + radius string
`placeObjectId`	string	Optional	—	Tweets tagged with a place
`minimumRetweets`	number	Optional	—	Minimum retweet count
`minimumFavorites`	number	Optional	—	Minimum like count
`minimumReplies`	number	Optional	—	Minimum reply count
`start`	string	Optional	—	Tweets after this date (YYYY-MM-DD)
`end`	string	Optional	—	Tweets before this date (YYYY-MM-DD)
`includeSearchTerms`	boolean	Optional	`false`	Add the matched search term to each tweet
`customMapFunction`	string	Optional	—	JavaScript function to transform each output object

Workflow

Progress:
- [ ] Step 1: Run scrapers for all three platforms in parallel
- [ ] Step 2: Classify sentiment per platform
- [ ] Step 3: Calculate cross-platform brand health score
- [ ] Step 4: Identify top themes and alerts
- [ ] Step 5: Deliver unified report

Step 1: Run Three Scrapers

Twitter (If Apify MCP is available):

Tool: apify:run-actor
Actor: "apidojo~tweet-scraper"
Input: {"searchTerms": ["[BRAND_NAME]"], "maxItems": 300, "tweetLanguage": "en"}

Reddit:

Tool: apify:run-actor
Actor: "apidojo~tweet-scraper"
Input: {"searches": ["[BRAND_NAME]"], "maxItems": 200, "sort": "new", "time": "month"}

TikTok:

Tool: apify:run-actor
Actor: "apidojo~tiktok-scraper"
Input: {"keywords": ["#[brandname]", "#[brandname]review"], "maxItems": 200}

REST API fallback — run each sequentially:

bash

# Twitter
curl -X POST "https://api.apify.com/v2/acts/apidojo~tweet-scraper/runs?token=$APIFY_TOKEN"   -H "Content-Type: application/json"   -d '{"searchTerms": ["[BRAND_NAME]"], "maxItems": 300}'

# Reddit
curl -X POST "https://api.apify.com/v2/acts/apidojo~tweet-scraper/runs?token=$APIFY_TOKEN"   -H "Content-Type: application/json"   -d '{"searches": ["[BRAND_NAME]"], "maxItems": 200, "sort": "new", "time": "month"}'

Step 2: Sentiment Classification

Use the same lexical model for all platforms (positive/negative/neutral indicators from analyzing-twitter-sentiment-for-topic skill). Weight by platform-specific engagement:

Twitter: likeCount + replyCount * 3
Reddit: upvotes + commentCount * 2
TikTok: playCount / 1000 + diggCount

Step 3: Brand Health Score

platform_sentiment[p] = (positive_count[p] - negative_count[p]) / total_count[p]  # range: -1 to +1

platform_weight = {twitter: 0.35, reddit: 0.40, tiktok: 0.25}  # Reddit = most considered opinion

brand_health_score = sum(platform_sentiment[p] * platform_weight[p] for p in platforms)
brand_health_score = (brand_health_score + 1) / 2 * 100  # normalize to 0-100

Score interpretation: 0–40 = Crisis, 40–55 = Concerning, 55–70 = Neutral, 70–85 = Positive, 85–100 = Strong.

Step 4: Edge Cases

Brand name is a common word (e.g. "Apple"): Add qualifier ("Apple iPhone", "Apple Inc") to search to reduce noise; report disambiguation rate
One platform dominates volume (e.g. TikTok has 10× Twitter posts): Weight by volume in the composite score
Rapid sentiment shift (score changes > 20 points): Flag as ALERT — may indicate PR crisis or viral positive moment
Reddit returns no results: Brand may not be discussed there; set reddit_weight = 0 and redistribute to other platforms

Output Format

# Cross-Platform Brand Sentiment: [BRAND_NAME]
Period: [DATE_RANGE] | Total posts: [N] | Date: [DATE]

## Brand Health Score: [X]/100 — [INTERPRETATION]

## Per-Platform Breakdown
| Platform | Posts | Positive | Negative | Neutral | Score |
|----------|-------|----------|----------|---------|-------|
| Twitter | [N] | [X%] | [X%] | [X%] | [+/-X] |
| Reddit | [N] | [X%] | [X%] | [X%] | [+/-X] |
| TikTok | [N] | [X%] | [X%] | [X%] | [+/-X] |

## Top Negative Themes (Cross-Platform)
1. [Theme] — [N] posts across [platforms]
2. [Theme]

## Top Positive Themes
1. [Theme] — [N] posts
2. [Theme]

## Most Impactful Posts
🔴 Top negative: [platform] | [handle] | [N engagement] | "[excerpt]"
🟢 Top positive: [platform] | [handle] | [N engagement] | "[excerpt]"

Troubleshooting

Brand health score conflicts between platforms: This is meaningful signal — discuss in output why platforms diverge (e.g. "Reddit community discusses product quality issues while TikTok shows positive unboxing content"). Sample too small for reliable sentiment (< 50 posts per platform): Widen date range or note low confidence in that platform's score. Brand name not found on a platform: Some brands have no organic TikTok presence — note as gap in output.

Recommended — run_actor.js (handles waiting, output, and file saving automatically):

bash

# Quick answer (prints table to chat)
node scripts/run_actor.js \
  --actor "apidojo~tweet-scraper" \
  --input '{"param": "value"}'

# Save as CSV
node scripts/run_actor.js \
  --actor "apidojo~tweet-scraper" \
  --input '{"param": "value"}' \
  --output YYYY-MM-DD_results.csv --format csv

# Save as JSON
node scripts/run_actor.js \
  --actor "apidojo~tweet-scraper" \
  --input '{"param": "value"}' \
  --output YYYY-MM-DD_results.json --format json

APIFY_TOKEN must be set in environment or .env file.

tracking-brand-sentiment-across-platforms