Latent Semantic Mapping
- Exclusive ContentScan the cover in the app
- CommunityDiscuss & Share
About this book
Latent semantic mapping (LSM) extends latent semantic analysis (LSA), a method first built to uncover hidden patterns among the words in a collection of documents. Traditional keyword matching often misses what a searcher means, so LSA aims to match queries to documents by underlying concepts. Its starting premise is that meaningful structure lies beneath the data but is blurred by the unpredictable ways people choose their words. Algebraic and statistical tools estimate that structure and filter out the noise. The result is a compact, continuous set of numerical values for words and documents, which takes the place of the original representation for indexing and retrieval.
The approach rests on three features: discrete items such as words and documents are placed in a continuous vector space, that placement is shaped by correlations found across the whole collection, and reducing the number of dimensions is built into the process. Because these properties are general, the author argues the paradigm applies well beyond text. LSM is presented as a data-driven framework for modeling global relationships in large bodies of data, not necessarily written language.
This monograph offers a general overview of the framework and shows how it can help with problems in natural language understanding and spoken language processing. It closes by weighing the trade-offs of the method and considering how far it can support data-driven information extraction.
Reader reviews
No reviews yet. Share your thoughts on this book in the Appilog app.
Book Details
- ISBN
- 9781598291049
- Author
- Jerome R. Bellegarda
- Publisher
- Unknown Publisher
- Genres
- Technology & ComputingLanguage Learning & Linguistics
Media & Extras
This book has 0 videos that play on its cover. Scan the cover with the Appilog app to watch.
Get the app
