Basic algorithm for nearest-neighb......

Tour Communities Teachers Enterprises

Help

Learning Communities > Data Mining > Text Mining - Predictive Methods by Weiss flashcards

Flashcard (Front and Back)

Front (Question)

Basic algorithm for nearest-neighbor algorithm for documents

Back (Answer)

1.) compute similarity of new document to all documents in collection

2.) Select k documents that are most similar to new document.

3.) Category is label that occurs most frequently in k selected documents.

Memory Work Out Sessions use brain research to accelerate and improve memorization. Automatically scheduled daily review drills ensure sure you don't forget.

or Find more than 100,000 other things to learn

Educators and Trainers:

Tour YoYoBrain's resources for learning and teaching

Previous Card

See the whole card set

Next Card

Current rating: 0.0/5
1
2
3
4
5

rated by 0 users

Comments:

Other flashcards from Text Mining - Predictive Methods by Weiss
Basic way to turn text document into ... What is a primary characteristic of d... What term is used for collections of ... Typical first step with document coll... Definition: stemming Define: inflectional stemming What is a way to reduce complexity of... Define: tf-idf formulation for weight... Since the numerical grid of word appe... Define: multiword features in text ... Define: information gain criterion What is the best known algorithm for ... Define: text categorization 4 common ways of text categorization Basic algorithm for nearest-neighbor ... Text Mining: define - cosine similarity Basic principle of decision rule anal... Use of linear scoring in text categor... What type of text causes severe probl... Define: hinge loss function Define: numerically ill-posed Define: regularization of linear cla... Contrast: recall and precision of tex... Contrast: recall and precision of te... Define: coreference resolution