**Term Frequency–Inverse Document Frequency** — the classic information retrieval formula still used in BM25 and hybrid search.
TF(t, d) = count(t in d) / len(d)IDF(t, D) = log(|D| / df(t))TF-IDF = TF × IDFWhere df(t) = number of documents containing term t.
term = "python" doc = ["python", "is", "great"] corpus = [["python","is","great"], ["java","is","fast"]] → TF = 1/3, IDF = log(2/1) = 0.69315 → TF-IDF = 0.23105
Round to **5 decimal places**.
Similar Problems
Test Cases (2 visible · 1 hidden)
Case 1: Basic TF-IDF
Input: tfidf("python",["python","is","great"],[["python","is","great"],["java","is","fast"]])
Expected: 0.23105
Case 2: Appears in all docs IDF=0
Input: tfidf("is",["python","is","great"],[["python","is","great"],["java","is","fast"]])
Expected: 0.0
⌘↵ Run · ⌘⇧↵ Submit