Topic vector
A numerical vector encoding the strengths of associations between terms or documents and latent topics.
Natural Language Processing in Action, Second Edition
A topic vector is a numerical representation whose components encode the strength or proportion of topics associated with a document.
- How it is used
- Represents words and documents for semantic search, document comparison, and classification using topic scores instead of word-by-word features.
- How deep it goes
- Develops a weighted three-topic thought experiment, then computes and evaluates learned LDiA vectors; the hand-picked weights are not a learning algorithm.
- Examples
- Uses pet-and-city topic weights and a 16-topic comment model evaluated for toxicity classification.
- What you can do after
- Apply it via scikit-learn to transform comment counts into topic vectors and evaluate them as classifier features.
Read how Natural Language Processing in Action, Second Edition explains topic vector →
Math and Architectures of Deep Learning
A vector whose entries indicate how strongly each term contributes to a topic.
- How it is used
- Represents latent topics and supports document comparison after reducing document-term vectors to topic space.
- How deep it goes
- Explains term contributions, eigenvector construction, topic weighting, cutoff selection, projection, and cosine similarity.
- Examples
- Uses a toy document-term matrix containing violence, gun, America, roses, and documents d0 through d6.
- What you can do after
- Evaluate document relationships by projecting term vectors into a retained topic space and comparing them with cosine similarity.
Read how Math and Architectures of Deep Learning explains topic vector →
Related concepts
- latent Dirichlet allocationproduced by · Latent Dirichlet allocation is a generative probabilistic topic model representing documents as mixtures of latent topics and topics as distributions over words.
- term frequency-inverse document frequencyproduced by · Term frequency-inverse document frequency is a text-weighting scheme that combines a term’s frequency in a document with its inverse frequency across a document collection.
- vectorkind of · A vector is an ordered, indexed sequence of elements, often numerical, represented as a mathematical quantity or a computational collection.
- cosine distancemeasured by · Cosine distance is a measure of dissimilarity between nonzero vectors based on their angular separation, commonly defined as one minus cosine similarity.
- latent semantic analysisproduced by · Latent semantic analysis is a dimensionality-reduction technique that factorizes term-document statistics to represent words and documents in a lower-dimensional semantic space.
- similarity searchuses · Similarity search is a retrieval technique that finds or ranks items by comparing their representations with a query according to a similarity or distance measure.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.