Lexical paraphrasing is the task of extracting word-level paraphrases. Lexical paraphrases should be context dependent since a word may have different paraphrases in distinct contexts. This paper investigates a framework for acquiring context-dependent lexical paraphrases, in which a web mining method is developed for extracting candidate paraphrases and a classification method is introduced in paraphrase validation. Evaluations are carried out on the People’s Daily corpus and the results show that: (1) the web mining method performs well in candidate paraphrase extraction, which extracts 2.3 correct paraphrases on average for each test word in each given context sentence; (2) the classifier for paraphrase validation is effective, which achieves an f-measure of 0.6023; (3) 75.11% and 98.31% of the paraphrases extracted by our method cannot be recognized by the two widely used context-independent methods, i.e., the thesaurus-based and clustering-based methods respectively. This indicates that the presented context-dependent method is a considerable supplement to the context-independent ones.