Text Sentiment Analyzer

Score the sentiment of any text — positive, negative, or neutral — with no API key and no setup.

Live on the Apify Store — run it instantly or call it as an agent tool via Apify MCP.

What you get

• A clear label: positive, negative, or neutral
• A raw score and a length-normalized comparative score
• The exact positive and negative words that drove the result
• Single text or bulk arrays in one run — each scored independently
• Fast, deterministic, lexicon-based (AFINN) — no model calls, no rate limits, no key

Input

{
  "text": "I absolutely love this product, it works wonderfully — though shipping was a bit slow.",
  "texts": ["Worst purchase ever, totally broken.", "Decent enough, nothing special."],
  "maxTexts": 50
}

Provide text, texts, or both.

Output

One dataset item per text:

{
  "text": "I absolutely love this product, it works wonderfully — though shipping was a bit slow.",
  "label": "positive",
  "score": 6,
  "comparative": 0.4615,
  "positiveWords": ["love", "wonderfully"],
  "negativeWords": ["slow"],
  "tokenCount": 13
}

Pricing

$0.01 per text analyzed — billed as the text-analyzed event. Bulk runs are capped by maxTexts, which is also your budget cap.

API & AI-agent access

Score text from any HTTP client and get the results in the same call:

curl -X POST "https://api.apify.com/v2/acts/eliai~text-sentiment-analyzer/run-sync-get-dataset-items?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"texts": ["Great support, terrible shipping."]}'

AI agents can call this actor as a tool through Apify's MCP server — deterministic scoring makes it a safe, cheap pre-filter before spending LLM tokens.

FAQ

How do I analyze the sentiment of customer reviews in bulk? Put them in texts (each scored independently), run, and sort the dataset by label or comparative — the length-normalized score that makes short and long texts comparable.

Why does it return the driver words? positiveWords and negativeWords show exactly which words produced the score, so every result is auditable — no black box, and mislabeled edge cases are visible instead of silent.

Is this an AI model? No — deliberately. It's lexicon-based (AFINN): deterministic, instant, no rate limits, no API key, and the same input always scores the same. For nuanced sarcasm you'd want an LLM; for high-volume filtering and trend tracking, deterministic and $0.01 wins.

What languages does it support? English — the AFINN lexicon is English-only. Non-English text scores near neutral rather than erroring.

What do score and comparative mean? score is the raw sum of word valences; comparative divides by token count, so a short angry sentence and a long mixed review can be ranked fairly on the same scale.

For AI agents

This Actor is built to be called by software, not just by people.

• Mount it directly as an MCP tool — no Store search, no ranking, just this one tool: https://mcp.apify.com/?actors=eliai/text-sentiment-analyzer
• Or call it over HTTP and get the results in the same request: POST https://api.apify.com/v2/acts/eliai~text-sentiment-analyzer/run-sync-get-dataset-items
• Pay with x402, without an Apify account. This Actor is whitelisted for agentic payments, so an agent holding USDC on Base can buy a prepaid token and spend it here. The minimum purchase is $1, the token balance is an absolute spending cap, and it expires 14 days after purchase.
• Costs are predictable before you call. Pricing is pay-per-event (see Pricing above), so an agent can budget a run in advance instead of discovering the bill afterwards.
• Send only the field you mean. If you pass the bulk field, it is used on its own; the single-value field is a fallback, never merged into your request. You are charged for the items you sent and nothing else.