One challenge for relevance ranking in Web search is underspecified queries. For such queries, top-ranked documents may contain information irrelevant to the search goal of the user; some newly-created relevant documents are ranked lower due to their freshness and to the large number of existing documents that match the queries. To improve the relevance ranking for underspecified queries requires better understanding of users' search goals. By analyzing the semantic query context extracted from the query logs, we propose Q-Rank to effectively improve the ranking of search results for a given query. Experiments show that Q-Rank outperforms the current ranking system of a large-scale commercial Web search engine, improving the relevance ranking for 82% of the queries with an average increase of 8.99% in terms of discounted cumulative gains. Because Q-Rank is independent of the underlying ranking algorithm, it can be integrated with existing search engines.