Explain how 'hybrid search methods' combining keyword and semantic search address limitations inherent in either method alone for comprehensive retrieval.

Hybrid search methods combine keyword-based search and semantic search to overcome the inherent limitations of each when used independently, thereby achieving more comprehensive and relevant retrieval. Keyword search, also known as lexical search, operates by matching exact terms or their variations within documents. Its strength lies in its precision when users know the exact words to search for, quickly identifying documents containing those specific terms. However, its primary limitation is the 'lexical gap': it struggles with synonyms (e.g., searching for 'automobile' will miss documents using 'car'), related concepts, misspellings, or when a user's query uses different phrasing than the document content. It lacks an understanding of the underlying meaning or context of words, leading to missed relevant documents if the exact keywords are not present. Semantic search, conversely, focuses on understanding the meaning and contextual relevance of queries and documents. It uses techniques like word embeddings and vector similarity to represent words and phrases as numerical vectors in a high-dimensional space, where semantically similar items are closer together. This allows it to find documents that are conceptually related to a query, even if they don't share exact keywords. For example, searching for 'fast vehicle' could return documents mentioning 'sports car'. Its strengths include handling natural language queries, synonyms, and conceptual relationships, improving recall by finding broader sets of relevant information. However, semantic search can sometimes lack precision for very specific queries, potentially returning conceptually similar but less directly relevant results, and can be computationally more intensive. Hybrid search methods address these limitations by fusing the two approaches. They leverage the precision of keyword search to ensure highly relevant, direct matches when exact terms are important, acting as an anchor for literal intent. Simultaneously, they employ semantic search to capture the nuanced meaning, context, and conceptual relationships, overcoming the lexical gap of keyword search. For instance, a query for 'Python' might use keyword search to prioritize documents containing 'Python programming language' while using semantic search to also find documents discussing 'data science libraries' or 'scripting languages', even if the exact word 'Python' isn't predominant. This combination improves recall by finding semantically relevant documents that keyword search would miss, and enhances precision by ensuring that exact term matches are highly ranked, preventing the conceptual drift sometimes seen in pure semantic search. The typical process involves performing both keyword and semantic searches, then combining or re-ranking their results based on a weighted relevance score, ensuring both broad conceptual coverage and specific lexical accuracy.