100 크레딧 무료, 카드 등록 없이지금 시작하기
Logo
Content Analysis logo
100 크레딧 무료, 카드 등록 없이

Content Analysis Search API

API 한 번의 호출로 Content Analysis Search 데이터를 받아 가세요. Returns paginated web citations (news, blogs, ecommerce, message-boards, organization pages) that mention the keyword, each scored on a 6-axis emotional sentiment model (anger, happiness, love, sadness, share, fun) plus positive/negative/neutral connotation probabilities. This is web-wide brand listening over a web-wide citation index: page-level mentions, not native platform posts. Cursor-paginated via `cursor`; `total` is the global match count, not the page size.

2026년 10월 업데이트SocialCrawl 팀이 직접 관리해요

키워드를 언급한 뉴스, 블로그, 쇼핑몰, 게시판 페이지를 감성 점수와 감정 톤까지 매겨 반환합니다.

키워드당 집계 하나만 필요하면 summary가 맞습니다. 개별 언급을 직접 읽고 페이징할 때 사용하세요.

69개 플랫폼을 병렬로 검색합니다

·TikTok logoTikTok·Instagram logoInstagram·YouTube logoYouTube·Facebook logoFacebook·X logoX·LinkedIn logoLinkedIn·Reddit logoReddit·Threads logoThreads·Pinterest logoPinterest·Twitch logoTwitch·Truth Social logoTruth Social·Snapchat logoSnapchat·Kick logoKick·Bluesky logoBluesky·Kwai logoKwai·Rumble logoRumble·Spotify logoSpotify·Apple Music logoApple Music·TikTok Shop logoTikTok Shop·Amazon Shop logoAmazon Shop·Google Shopping logoGoogle Shopping·Trustpilot logoTrustpilot·TripAdvisor logoTripAdvisor·Yelp logoYelp·Linktree logoLinktree·Komi logoKomi·Pillar logoPillar·lnk.bio logolnk.bio·Facebook Ads logoFacebook Ads·Google Ads logoGoogle Ads·LinkedIn Ads logoLinkedIn Ads·Google Search logoGoogle Search·Google News logoGoogle News·Finance logoFinance·Polymarket logoPolymarket·Tavily logoTavily·Hacker News logoHacker News·GitHub logoGitHub·Perplexity logoPerplexity·Naver logoNaver·Utility logoUtility·Universal Search logoUniversal Search
·TikTok logoTikTok·Instagram logoInstagram·YouTube logoYouTube·Facebook logoFacebook·X logoX·LinkedIn logoLinkedIn·Reddit logoReddit·Threads logoThreads·Pinterest logoPinterest·Twitch logoTwitch·Truth Social logoTruth Social·Snapchat logoSnapchat·Kick logoKick·Bluesky logoBluesky·Kwai logoKwai·Rumble logoRumble·Spotify logoSpotify·Apple Music logoApple Music·TikTok Shop logoTikTok Shop·Amazon Shop logoAmazon Shop·Google Shopping logoGoogle Shopping·Trustpilot logoTrustpilot·TripAdvisor logoTripAdvisor·Yelp logoYelp·Linktree logoLinktree·Komi logoKomi·Pillar logoPillar·lnk.bio logolnk.bio·Facebook Ads logoFacebook Ads·Google Ads logoGoogle Ads·LinkedIn Ads logoLinkedIn Ads·Google Search logoGoogle Search·Google News logoGoogle News·Finance logoFinance·Polymarket logoPolymarket·Tavily logoTavily·Hacker News logoHacker News·GitHub logoGitHub·Perplexity logoPerplexity·Naver logoNaver·Utility logoUtility·Universal Search logoUniversal Search
·TikTok logoTikTok·Instagram logoInstagram·YouTube logoYouTube·Facebook logoFacebook·X logoX·LinkedIn logoLinkedIn·Reddit logoReddit·Threads logoThreads·Pinterest logoPinterest·Twitch logoTwitch·Truth Social logoTruth Social·Snapchat logoSnapchat·Kick logoKick·Bluesky logoBluesky·Kwai logoKwai·Rumble logoRumble·Spotify logoSpotify·Apple Music logoApple Music·TikTok Shop logoTikTok Shop·Amazon Shop logoAmazon Shop·Google Shopping logoGoogle Shopping·Trustpilot logoTrustpilot·TripAdvisor logoTripAdvisor·Yelp logoYelp·Linktree logoLinktree·Komi logoKomi·Pillar logoPillar·lnk.bio logolnk.bio·Facebook Ads logoFacebook Ads·Google Ads logoGoogle Ads·LinkedIn Ads logoLinkedIn Ads·Google Search logoGoogle Search·Google News logoGoogle News·Finance logoFinance·Polymarket logoPolymarket·Tavily logoTavily·Hacker News logoHacker News·GitHub logoGitHub·Perplexity logoPerplexity·Naver logoNaver·Utility logoUtility·Universal Search logoUniversal Search
·TikTok logoTikTok·Instagram logoInstagram·YouTube logoYouTube·Facebook logoFacebook·X logoX·LinkedIn logoLinkedIn·Reddit logoReddit·Threads logoThreads·Pinterest logoPinterest·Twitch logoTwitch·Truth Social logoTruth Social·Snapchat logoSnapchat·Kick logoKick·Bluesky logoBluesky·Kwai logoKwai·Rumble logoRumble·Spotify logoSpotify·Apple Music logoApple Music·TikTok Shop logoTikTok Shop·Amazon Shop logoAmazon Shop·Google Shopping logoGoogle Shopping·Trustpilot logoTrustpilot·TripAdvisor logoTripAdvisor·Yelp logoYelp·Linktree logoLinktree·Komi logoKomi·Pillar logoPillar·lnk.bio logolnk.bio·Facebook Ads logoFacebook Ads·Google Ads logoGoogle Ads·LinkedIn Ads logoLinkedIn Ads·Google Search logoGoogle Search·Google News logoGoogle News·Finance logoFinance·Polymarket logoPolymarket·Tavily logoTavily·Hacker News logoHacker News·GitHub logoGitHub·Perplexity logoPerplexity·Naver logoNaver·Utility logoUtility·Universal Search logoUniversal Search
Content Analysis API

Search API로 무엇을 할 수 있을까요

Search 엔드포인트가 통합 스키마와 계산 필드를 담은 Content Analysis 데이터를 한 번의 요청으로 보내드려요. 스크래핑 인프라를 직접 만들거나 유지할 필요가 없어요.

요청 예시

# GET /v1/content_analysis/search: 20 credits per call. Cache hits are free and failed calls are refunded.
# Check the price first for 0 credits: add dry_run=1 to the query. Nothing is fetched or charged.
# --max-time 60: the API's own time limit for this endpoint.
# --retry 3: retries timeouts, 429 and 5xx, honoring Retry-After.
# -o response.json keeps only the final attempt's body.
# One Idempotency-Key per request, reused on every retry, so a retried call is never charged twice.
IDEMPOTENCY_KEY=$(od -An -N16 -tx1 /dev/urandom | tr -d ' \n')
curl -sS --max-time 60 --retry 3 -o response.json \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Idempotency-Key: $IDEMPOTENCY_KEY" \
  "https://www.socialcrawl.dev/v1/content_analysis/search?keyword=openai&search_mode=as_is&limit=10" && cat response.json
# GET /v1/content_analysis/search: 20 credits per call. Cache hits are free and failed calls are refunded.
# Check the price first for 0 credits: add dry_run=1 to the query. Nothing is fetched or charged.
import os
import random
import time
import uuid

import requests

API_KEY = os.environ.get("SOCIALCRAWL_API_KEY", "YOUR_API_KEY")
TIMEOUT_S = 60  # the API's own time limit for this endpoint.
MAX_RETRIES = 3
URL = "https://www.socialcrawl.dev/v1/content_analysis/search"


def backoff(attempt):
    # Exponential backoff with jitter: about 1 s, 2 s, then 4 s.
    return 2 ** attempt * (0.5 + random.random() / 2)


def socialcrawl(params):
    # One Idempotency-Key per request, reused on every retry, so a retried call is never charged twice.
    idempotency_key = str(uuid.uuid4())
    for attempt in range(MAX_RETRIES + 1):
        try:
            res = requests.get(
                URL,
                params=params,
                headers={"x-api-key": API_KEY, "Idempotency-Key": idempotency_key},
                timeout=TIMEOUT_S,
            )
            body = res.json()
        except (requests.RequestException, ValueError):
            # A timeout, a dropped connection or a non-JSON gateway page: retry with the same key.
            if attempt == MAX_RETRIES:
                raise
            time.sleep(backoff(attempt))
            continue
        if body["success"]:
            return body
        error = body["error"]
        # Branch on error["type"], never on the message text.
        if error["type"] == "RESOURCE_NOT_FOUND":
            return None  # refunded: nothing to read
        retry_after = res.headers.get("Retry-After")
        # SERVICE_UNAVAILABLE without Retry-After is a withdrawn endpoint: permanent.
        permanent = error["type"] == "SERVICE_UNAVAILABLE" and retry_after is None
        if not error.get("retryable") or permanent or attempt == MAX_RETRIES:
            raise RuntimeError(f"{error['type']}: {error['message']} (request_id {body['request_id']})")
        # Retry-After (seconds) wins; otherwise back off.
        time.sleep(float(retry_after) if retry_after else backoff(attempt))


params = {
    "keyword": "openai",
    "search_mode": "as_is",
    "limit": "10",
}
body = socialcrawl(params)
if body is not None:
    print(body["data"])
// GET /v1/content_analysis/search: 20 credits per call. Cache hits are free and failed calls are refunded.
// Check the price first for 0 credits: add dry_run=1 to the query. Nothing is fetched or charged.
const API_KEY = process.env.SOCIALCRAWL_API_KEY ?? "YOUR_API_KEY";
const TIMEOUT_MS = 60_000; // the API's own time limit for this endpoint.
const MAX_RETRIES = 3;

type Page = { next_cursor: string | null; has_more: boolean };
type SocialCrawlResponse =
  | { success: true; data: any; pagination?: Page; credits_used: number; request_id: string }
  | { success: false; error: { type: string; message: string; retryable: boolean }; request_id: string };

const sleep = (ms: number) => new Promise((resolve) => setTimeout(resolve, ms));
// Exponential backoff with jitter: about 1 s, 2 s, then 4 s.
const backoffMs = (attempt: number) => 2 ** attempt * 1000 * (0.5 + Math.random() / 2);

async function socialcrawl(url: URL) {
  // One Idempotency-Key per request, reused on every retry, so a retried call is never charged twice.
  const idempotencyKey = crypto.randomUUID();
  for (let attempt = 0; ; attempt++) {
    let res: Response;
    let body: SocialCrawlResponse;
    try {
      res = await fetch(url, {
        headers: {
          "x-api-key": API_KEY,
          "Idempotency-Key": idempotencyKey,
        },
        signal: AbortSignal.timeout(TIMEOUT_MS),
      });
      body = (await res.json()) as SocialCrawlResponse;
    } catch (error) {
      // A timeout, a dropped connection or a non-JSON gateway page: retry with the same key.
      if (attempt >= MAX_RETRIES) throw error;
      await sleep(backoffMs(attempt));
      continue;
    }
    if (body.success) return body;
    // Branch on error.type, never on the message text.
    if (body.error.type === "RESOURCE_NOT_FOUND") return null; // refunded: nothing to read
    const retryAfter = res.headers.get("Retry-After");
    // SERVICE_UNAVAILABLE without Retry-After is a withdrawn endpoint: permanent.
    const permanent = body.error.type === "SERVICE_UNAVAILABLE" && retryAfter === null;
    if (!body.error.retryable || permanent || attempt >= MAX_RETRIES) {
      throw new Error(`${body.error.type}: ${body.error.message} (request_id ${body.request_id})`);
    }
    // Retry-After (seconds) wins; otherwise back off.
    await sleep(retryAfter ? Number(retryAfter) * 1000 : backoffMs(attempt));
  }
}

const url = new URL("https://www.socialcrawl.dev/v1/content_analysis/search");
url.searchParams.set("keyword", "openai");
url.searchParams.set("search_mode", "as_is");
url.searchParams.set("limit", "10");

const body = await socialcrawl(url);
if (body) console.log(body.data);

파라미터

파라미터필수설명
keyword예Brand or term to find mentions of. Wrap in escaped double-quotes for an exact phrase (e.g. "logitech mouse").
page_type아니오Narrow to one or more page types (comma-separated): ecommerce, news, blogs, message-boards, organization. Translated to a page_types filter upstream. (ecommerce | news | blogs | message-boards | organization)
search_mode아니오as_is (default) returns every matching page; one_per_domain dedupes to the top page per domain. (as_is | one_per_domain)
limit아니오Number of citations to return per page (1-100, default 10). Paginate via `cursor` for more.
cursor아니오Opaque pagination cursor: pass the `next_cursor` from the previous response to fetch the next page.
order_by아니오Sort rules as "field,direction"; separate multiple rules with ";" (e.g. content_info.sentiment_connotations.anger,desc).
filters아니오Advanced filter expression as a JSON array (≤8 conditions). Combined with page_type via AND when both are present.
응답 예시

Content Analysis Search API는 무엇을 돌려주나요

모든 응답은 하나의 통합 스키마를 따라요. 크레딧을 쓰기 전에 어떤 필드가 돌아오는지, 실제 응답 본문 그대로 확인해 보세요.

응답 예시 보기
{
  "success": true,
  "platform": "content_analysis",
  "endpoint": "/v1/content_analysis/search",
  "data": {
    "items": [
      {
        "url": "https://timesofindia.indiatimes.com/topic/openai",
        "domain": "timesofindia.indiatimes.com",
        "main_domain": "indiatimes.com",
        "country": "IN",
        "language": "en",
        "domain_rank": 648,
        "spam_score": 0,
        "fetch_time": "2026-09-26 14:41:03 +00:00",
        "page_types": [
          "news",
          "organization"
        ],
        "title": "OpenAI",
        "snippet": "OpenAI is a private artificial intelligence research group created by Sam Altman, Greg Brockman, Ilya Sutskever, Wojciech Zaremba, and Elon Musk in December 2015.\nThe fundamental point of establishing OpenAI was to responsibly develop and promote benign AI. The creators believed that AI had the ability to aid in the resolution of some of the world's most urgent issues.\nStarting off as a research organization in the field of artificial intelligence, it eventually developed GPT-2, a language model capable of producing human-like prose, as well as OpenAI Gym, a toolset for designing and comparing reinforcement learning algorithms.\nAside from research and development, OpenAI is involved in policy work, education and outreach, and the development of AI applications. The company also has a number of agreements with other AI-related organizations, including Microsoft, Google, and IBM.\nDALL-E, another AI model, was one of OpenAI's core efforts. It is a generative model capable of producing graphics from text descriptions. It is based on the GPT-3 model, which is one of OpenAI's most advanced language models.\nThe board of directors of OpenAI removed Sam Altman, the organization's co-founder and CEO. The principal reason given for his dismissal was a lack of constant openness in his discussions with board members, which was considered as impeding their ability to carry out their obligations efficiently.\nThis decision was the result of a rigorous evaluation process by the board, which resulted in a lack of trust in Altman's capacity to lead OpenAI in the future.\nMira Murati is the new interim CEO and CTO of OpenAI. Murati is also one of the leading developers of ChatGPT.",
        "sentiment": {
          "anger": 0.06283660978078842,
          "happiness": 0.27491387724876404,
          "love": 0.15093187987804413,
          "sadness": 0.0974915474653244,
          "share": 0.234482541680336,
          "fun": 0.17940358817577362
        },
        "connotation": {
          "positive": 0.31331169605255127,
          "negative": 0.4118451774120331,
          "neutral": 0.27487316727638245
        },
        "rating": null,
        "text_category": [
          10108,
          10013
        ],
        "date_published": null,
        "_raw": {
          "type": "content_analysis_search",
          "url": "https://timesofindia.indiatimes.com/topic/openai",
          "domain": "timesofindia.indiatimes.com",
          "main_domain": "indiatimes.com",
          "url_rank": 368,
          "spam_score": 0,
          "domain_rank": 648,
          "fetch_time": "2026-09-26 14:41:03 +00:00",
          "country": "IN",
          "language": "en",
          "score": 534206.06,
          "page_category": [
            10108,
            10584
          ],
          "page_types": [
            "news",
            "organization"
          ],
          "ratings": null,
          "social_metrics": null,
          "content_info": {
            "content_type": "page_content",
            "title": "OpenAI",
            "main_title": "OpenAI",
            "previous_title": null,
            "level": 1,
            "author": null,
            "snippet": "OpenAI is a private artificial intelligence research group created by Sam Altman, Greg Brockman, Ilya Sutskever, Wojciech Zaremba, and Elon Musk in December 2015.\nThe fundamental point of establishing OpenAI was to responsibly develop and promote benign AI. The creators believed that AI had the ability to aid in the resolution of some of the world's most urgent issues.\nStarting off as a research organization in the field of artificial intelligence, it eventually developed GPT-2, a language model capable of producing human-like prose, as well as OpenAI Gym, a toolset for designing and comparing reinforcement learning algorithms.\nAside from research and development, OpenAI is involved in policy work, education and outreach, and the development of AI applications. The company also has a number of agreements with other AI-related organizations, including Microsoft, Google, and IBM.\nDALL-E, another AI model, was one of OpenAI's core efforts. It is a generative model capable of producing graphics from text descriptions. It is based on the GPT-3 model, which is one of OpenAI's most advanced language models.\nThe board of directors of OpenAI removed Sam Altman, the organization's co-founder and CEO. The principal reason given for his dismissal was a lack of constant openness in his discussions with board members, which was considered as impeding their ability to carry out their obligations efficiently.\nThis decision was the result of a rigorous evaluation process by the board, which resulted in a lack of trust in Altman's capacity to lead OpenAI in the future.\nMira Murati is the new interim CEO and CTO of OpenAI. Murati is also one of the leading developers of ChatGPT.",
            "snippet_length": 1688,
            "social_metrics": null,
            "highlighted_text": null,
            "language": "en",
            "sentiment_connotations": {
              "anger": 0.06283660978078842,
              "happiness": 0.27491387724876404,
              "love": 0.15093187987804413,
              "sadness": 0.0974915474653244,
              "share": 0.234482541680336,
              "fun": 0.17940358817577362
            },
            "connotation_types": {
              "positive": 0.31331169605255127,
              "negative": 0.4118451774120331,
              "neutral": 0.27487316727638245
            },
            "text_category": [
              10108,
              10013
            ],
            "date_published": null,
            "content_quality_score": 93,
            "semantic_location": null,
            "rating": null,
            "group_date": "2023-10-16 16:50:13 +00:00"
          }
        }
      },
      {
        "url": "https://plus.freshnewsasia.com/index.php/en/freshnewsplus/388133-2025-05-17-08-46-45.html",
        "domain": "plus.freshnewsasia.com",
        "main_domain": "freshnewsasia.com",
        "country": null,
        "language": "en",
        "domain_rank": 394,
        "spam_score": 0,
        "fetch_time": "2025-06-11 01:53:37 +00:00",
        "page_types": [
          "cms",
          "news"
        ],
        "title": "OpenAI គ្រោងជួយប្រទេសអេមីរ៉ាត់អារ៉ាប់រួមអភិវឌ្ឍមជ្ឈមណ្ឌលទិន្នន័យដ៏ធំបំផុតមួយរបស់ពិភពលោក",
        "snippet": "(បរទេស)៖ សារព័ត៌មាន Bloomberg បានរាយការណ៍នៅថ្ងៃសុក្រសប្តាហ៍នេះថា ក្រុមហ៊ុនបច្ចេកវិទ្យាយក្ស OpenAI គ្រោង​នឹង​ជួយ​អភិវឌ្ឍ​មជ្ឈមណ្ឌល​ទិន្នន័យ​ដ៏​ធំ​មួយ​នៅ​ទីក្រុង Abu Dhabi ប្រទេស​អារ៉ាប់រួម ដែល​អាច​ក្លាយ​ជា​មជ្ឈមណ្ឌល​ធំ​បំផុត​មួយ​ក្នុង​ពិភពលោក។ នេះបើយោងតាមការចេញផ្សាយដោយទីភ្នាក់ងារសារព័ត៌មាន Reuters នៅថ្ងៃទី១៧ ខែឧសភា ឆ្នាំ២០២៥។\nគម្រោងនេះត្រូវបានគេរំពឹងថានឹងមានសមត្ថភាព ៥ ជីហ្គាវ៉ាត់ ដែលធ្វើឱ្យវាក្លាយជាការពង្រីកដ៏សំខាន់នៃមហិច្ឆតាហេដ្ឋារចនាសម្ព័ន្ធ AI សកលរបស់ OpenAI។ របាយការណ៍ពី Bloomberg ដដែលបានឱ្យដឹងទៀតថា ការចូលរួមរបស់ក្រុមហ៊ុន OpenAI នៅមិនទាន់ត្រូវបានបញ្ចប់នៅឡើយនោះទេ ប៉ុន្តែការប្រកាសជាផ្លូវការអាចនឹងមកដល់ក្នុងពេលដ៏ខ្លីខាងមុខនេះ។​ យ៉ាងណាមិញ ជុំវិញព័ត៌មានលម្អិតអំពីរបាយការណ៍នេះ OpenAI មិនទាន់មានការឆ្លើយតបលើសំណើរសុំការអត្ថាធិប្បាយនៅឡើយនោះទេ។\nគួរជម្រាប​ថា ប្រទេសអេមីរ៉ាត់អារ៉ាប់រួម (UAE) និងសហរដ្ឋអាមេរិកនៅថ្ងៃព្រហស្បតិ៍សប្តាហ៍នេះ បានចុះហត្ថលេខាលើកិច្ចព្រមព្រៀងដ៏សំខាន់មួយសម្រាប់ UAE ដើម្បីសាងសង់បរិវេណសាលាបញ្ញាសិប្បនិម្មិត (AI) ដ៏ធំបំផុតនៅខាងក្រៅសហរដ្ឋអាមេរិក ដែលមានទីតាំងនៅ Abu Dhabi។ រោងចក្រនេះ ត្រូវបានគេបង្ហើបឱ្យដឹងថានឹងគ្របដណ្តប់លើផ្ទៃដីប្រហែល ១០ ម៉ាយការ៉េ ហើយនឹងប្រើប្រាស់ថាមពលស្មើនឹងម៉ាស៊ីនរ៉េអាក់ទ័រនុយក្លេអ៊ែរចំនួន ៥ ដោយកាត់បន្ថយហេដ្ឋារចនាសម្ព័ន្ធ AI ដែលមានស្រាប់ដោយ OpenAI ឬដៃគូប្រកួតប្រជែងរបស់វា៕",
        "sentiment": {
          "anger": 0.14812691509723663,
          "happiness": 0.18087238073349,
          "love": 0.17585210502147677,
          "sadness": 0.15571703016757965,
          "share": 0.18157997727394104,
          "fun": 0.15791159868240356
        },
        "connotation": {
          "positive": 0.3091999888420105,
          "negative": 0.5366879105567932,
          "neutral": 0.15414205193519592
        },
        "rating": null,
        "text_category": [
          10108,
          10017
        ],
        "date_published": "2025-05-17 08:46:45 +00:00",
        "_raw": {
          "type": "content_analysis_search",
          "url": "https://plus.freshnewsasia.com/index.php/en/freshnewsplus/388133-2025-05-17-08-46-45.html",
          "domain": "plus.freshnewsasia.com",
          "main_domain": "freshnewsasia.com",
          "url_rank": 286,
          "spam_score": 0,
          "domain_rank": 394,
          "fetch_time": "2025-06-11 01:53:37 +00:00",
          "country": null,
          "language": "en",
          "score": 420719.28,
          "page_category": [
            10108,
            13647
          ],
          "page_types": [
            "cms",
            "news"
          ],
          "ratings": null,
          "social_metrics": null,
          "content_info": {
            "content_type": "page_content",
            "title": "OpenAI គ្រោងជួយប្រទេសអេមីរ៉ាត់អារ៉ាប់រួមអភិវឌ្ឍមជ្ឈមណ្ឌលទិន្នន័យដ៏ធំបំផុតមួយរបស់ពិភពលោក",
            "main_title": "OpenAI គ្រោងជួយប្រទេសអេមីរ៉ាត់អារ៉ាប់រួមអភិវឌ្ឍមជ្ឈមណ្ឌលទិន្នន័យដ៏ធំបំផុតមួយរបស់ពិភពលោក",
            "previous_title": null,
            "level": 2,
            "author": null,
            "snippet": "(បរទេស)៖ សារព័ត៌មាន Bloomberg បានរាយការណ៍នៅថ្ងៃសុក្រសប្តាហ៍នេះថា ក្រុមហ៊ុនបច្ចេកវិទ្យាយក្ស OpenAI គ្រោង​នឹង​ជួយ​អភិវឌ្ឍ​មជ្ឈមណ្ឌល​ទិន្នន័យ​ដ៏​ធំ​មួយ​នៅ​ទីក្រុង Abu Dhabi ប្រទេស​អារ៉ាប់រួម ដែល​អាច​ក្លាយ​ជា​មជ្ឈមណ្ឌល​ធំ​បំផុត​មួយ​ក្នុង​ពិភពលោក។ នេះបើយោងតាមការចេញផ្សាយដោយទីភ្នាក់ងារសារព័ត៌មាន Reuters នៅថ្ងៃទី១៧ ខែឧសភា ឆ្នាំ២០២៥។\nគម្រោងនេះត្រូវបានគេរំពឹងថានឹងមានសមត្ថភាព ៥ ជីហ្គាវ៉ាត់ ដែលធ្វើឱ្យវាក្លាយជាការពង្រីកដ៏សំខាន់នៃមហិច្ឆតាហេដ្ឋារចនាសម្ព័ន្ធ AI សកលរបស់ OpenAI។ របាយការណ៍ពី Bloomberg ដដែលបានឱ្យដឹងទៀតថា ការចូលរួមរបស់ក្រុមហ៊ុន OpenAI នៅមិនទាន់ត្រូវបានបញ្ចប់នៅឡើយនោះទេ ប៉ុន្តែការប្រកាសជាផ្លូវការអាចនឹងមកដល់ក្នុងពេលដ៏ខ្លីខាងមុខនេះ។​ យ៉ាងណាមិញ ជុំវិញព័ត៌មានលម្អិតអំពីរបាយការណ៍នេះ OpenAI មិនទាន់មានការឆ្លើយតបលើសំណើរសុំការអត្ថាធិប្បាយនៅឡើយនោះទេ។\nគួរជម្រាប​ថា ប្រទេសអេមីរ៉ាត់អារ៉ាប់រួម (UAE) និងសហរដ្ឋអាមេរិកនៅថ្ងៃព្រហស្បតិ៍សប្តាហ៍នេះ បានចុះហត្ថលេខាលើកិច្ចព្រមព្រៀងដ៏សំខាន់មួយសម្រាប់ UAE ដើម្បីសាងសង់បរិវេណសាលាបញ្ញាសិប្បនិម្មិត (AI) ដ៏ធំបំផុតនៅខាងក្រៅសហរដ្ឋអាមេរិក ដែលមានទីតាំងនៅ Abu Dhabi។ រោងចក្រនេះ ត្រូវបានគេបង្ហើបឱ្យដឹងថានឹងគ្របដណ្តប់លើផ្ទៃដីប្រហែល ១០ ម៉ាយការ៉េ ហើយនឹងប្រើប្រាស់ថាមពលស្មើនឹងម៉ាស៊ីនរ៉េអាក់ទ័រនុយក្លេអ៊ែរចំនួន ៥ ដោយកាត់បន្ថយហេដ្ឋារចនាសម្ព័ន្ធ AI ដែលមានស្រាប់ដោយ OpenAI ឬដៃគូប្រកួតប្រជែងរបស់វា៕",
            "snippet_length": 1209,
            "social_metrics": null,
            "highlighted_text": null,
            "language": "en",
            "sentiment_connotations": {
              "anger": 0.14812691509723663,
              "happiness": 0.18087238073349,
              "love": 0.17585210502147677,
              "sadness": 0.15571703016757965,
              "share": 0.18157997727394104,
              "fun": 0.15791159868240356
            },
            "connotation_types": {
              "positive": 0.3091999888420105,
              "negative": 0.5366879105567932,
              "neutral": 0.15414205193519592
            },
            "text_category": [
              10108,
              10017
            ],
            "date_published": "2025-05-17 08:46:45 +00:00",
            "content_quality_score": 70,
            "semantic_location": "main",
            "rating": null,
            "group_date": "2025-05-17 08:46:45 +00:00"
          }
        }
      }
    ],
    "total": 17239034,
    "next_cursor": "eyJDdXJyZW50T2Zmc2V0IjoxMDAsIlJhd1JlcXVlc3QiOnsia2V5d29yZCI6Im9wZW5haSIsInNlYXJjaF9tb2RlIjoiYXNfaXMiLCJsaW1pdCI6MTAwLCJyYW5rX3NjYWxlIjoib25lX3Rob3VzYW5kIiwiYWlkIjo4NzM2OSwidGFza19pZCI6IjEwMDIxMjQzLTg3MzYtMDQ2My0wMDAwLWZhNmZiN2U2ZjdmOCJ9LCJTZWFyY2hBZnRlckRhdGEiOnsiVmVyc2lvbiI6MSwiU2VhcmNoQWZ0ZXJWYWx1ZXMiOnsiX3Njb3JlIjozNDk1MzcuNiwidXJsX2hhc2giOiItNzQyNTQ2NzU4NjMwMTc5MjkwNyJ9LCJUb2tlblJlYWxPZmZzZXQiOjB9fQ==",
    "dropped": 0
  },
  "credits_used": 20,
  "request_id": "req_example000000",
  "cached": false,
  "pagination": {
    "next_cursor": "sc.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",
    "has_more": true,
    "page_size": 100
  },
  "credits_remaining": 9999,
  "source": "captured",
  "captured_at": "2026-10-02T12:44:01.942Z",
  "redacted": true
}

통합 응답 구조를 보여 주는 실제 샘플이에요. 필드 값은 조회한 대상에 따라 달라져요.

API 상세

Content Analysis Search API는 어떻게 동작하나요

API 키와 함께 GET 요청을 보내면, 통합 스키마와 계산 필드를 담은 깔끔한 JSON이 돌아와요.

메서드

GET

응답 형식

JSON

왜 SocialCrawl인가요

Content Analysis Search 데이터는 왜 SocialCrawl로 가져올까요

Content Analysis 데이터 추출에서 까다로운 부분은 SocialCrawl이 가려 드려요. 통합 스키마, AI 인리치먼트, 그리고 코드에 남지 않는 플랫폼 로직까지.

개발자 우선

소셜 미디어 데이터를 몇 초 만에 수집하는 방법

개발자를 위한 가장 빠른 소셜 미디어 스크래핑 API. 월간 활성 사용자 100억 명 이상을 포괄하는 69개 플랫폼에서 프로필, 게시물, 댓글, 분석 데이터를 수집하세요.

모든 플랫폼을 하나의 스키마로

동일한 응답 구조로 69개 플랫폼을 조회하세요. 연동은 한 번이면 충분합니다.

단순 수집을 넘어 계산된 필드 제공

엔드포인트가 해당 지표를 지원하고 계산에 필요한 원본 값이 있을 때, 정규화된 레코드에 engagement_rate, estimated_reach, content_category, language를 함께 담아 바로 활용할 수 있습니다.

코드 한 줄 쓰기 전에, 데이터부터

Visual Data Explorer에 URL만 붙여넣으면 결과 카드와 정형화된 테이블, CSV 내보내기를 바로 사용할 수 있습니다.

import requests

response = requests.get(
    'https://www.socialcrawl.dev/v1/tiktok/profile',
    params={'handle': 'charlidamelio'},
    headers={'x-api-key': 'sc_YOUR_API_KEY'}
)
data = response.json()
[ .JSON ]
{
  "success": true,
  "platform": "tiktok",
  "data": {
    "author": {
      "username": "charlidamelio",
      "followers": 152400000
    },
    "engagement": {
      "likes": 12400000000,
      "engagement_rate": 0.087
    },
    "metadata": {
      "language": "en",
      "content_category": "lifestyle"
    }
  }
}
+ 69개 플랫폼
자주 묻는 질문

자주 묻는 질문

API, 요금제, 기능에 대한 질문과 답변입니다.

문의하기
브랜드의 웹 멘션은 어떻게 가져오나요?
GET /v1/content_analysis/search에 keyword 파라미터로 브랜드나 검색어를 담아 호출하면 돼요. 해당 단어를 언급한 웹 페이지가 URL·도메인·국가·언어·제목·스니펫과 6축 감성, 긍정·부정·중립 분류와 함께 페이지네이션돼 와요.
검색 결과에는 어떤 필드가 있나요?
항목마다 url, domain·main_domain, 국가, 언어, domain_rank, page_types, 제목, 스니펫, 6축 감성(anger·happiness·love·sadness·share·fun), 긍정·부정·중립 분류, rating, text_category, 작성일이 담겨요. 모두 하나의 통합 JSON 형태로 와요.
검색 엔드포인트의 페이지네이션은 어떻게 동작하나요?
search는 cursor 기반이에요. cursor를 넘기고 응답의 next_cursor를 읽어 다음 페이지를 받아요. 더 받을 페이지가 없으면 next_cursor는 null이 아니라 아예 빠져요. total은 페이지 크기가 아니라 전체 매칭 수라서 수백만까지 갈 수 있어요.
멘션을 페이지 유형으로 거를 수 있나요?
네, page_type에 ecommerce·news·blogs·message-boards·organization을 CSV로 넘기면 그 유형으로 좁혀져요. search_mode, limit, order_by, 그리고 고급 filters 식까지 더해 결과 범위를 한층 더 조절하실 수 있어요.
브랜드 멘션 검색은 요금이 얼마인가요?
한 번 요청에 20 크레딧이 들어요(어드밴스드 티어). 한 페이지에 멘션이 몇 개 오든 동일해요. 새 계정에는 100 크레딧이 무료로 제공돼서, 신용카드 없이 키워드와 필터를 먼저 테스트해 볼 수 있어요.

AI에게 SocialCrawl을 물어보세요

Content Analysis Search 데이터, 가져올 준비 되셨어요?

API 키 받고 60초 안에 Content Analysis 데이터를 받아 가세요.

무료로 시작하기

🤖 AI 에이전트나 LLM이신가요? 이 페이지를 markdown으로 읽어보세요