Fosslim/fosslim

Issues

• add WASM target

  #33 opened 5 years ago by timgluz
  0

• add minHashing

  #32 opened 5 years ago by timgluz
  0

• research Kernelized LocalSensitiveHashing

  #30 opened 7 years ago by timgluz
  2

• Research LocalitySensitiveHashing

  #19 opened 7 years ago by timgluz
  2

• research FB's starSpace;

  #28 opened 7 years ago by timgluz
  0

• research DocTensors and PointWise mutual information

  #29 opened 7 years ago by timgluz
  0

• [research] combSum & ways to combine results of multiple model

  #27 opened 7 years ago by timgluz
  0

• compare AVG(score) & Error over validation set

  #26 opened 7 years ago by timgluz
  0

• Clean up compiler warning and remove TODOs

  #12 opened 7 years ago by timgluz
  1

• implement TermVectorModel with Cosine distance;

  #2 opened 7 years ago by timgluz
  2

• optimize comparison by selecting only docs with matching hash

  #25 opened 7 years ago by timgluz
  2

• Add matching with fingerprinting

  #15 opened 7 years ago by timgluz
  4

• Research Okapi 25

  #24 opened 7 years ago by timgluz
  0

• Refactor models to use common traits;

  #23 opened 7 years ago by timgluz
  0

• Do sensitivity analysis

  #22 opened 7 years ago by timgluz
  0

• Research Winnowing paper for better fingerprinting

  #21 opened 7 years ago by timgluz
  0

• Research Ninka & paper

  #20 opened 7 years ago by timgluz
  0

• Create rubygem that uses fosslim underneath and provides Ruby class for text-matching

  #18 opened 7 years ago by timgluz
  1

• Add LicenseMatcher class for end-users;

  #17 opened 7 years ago by timgluz
  1

• Add model persistance methods;

  #16 opened 7 years ago by timgluz
  1

• optimize Document matching and add benchmarks;

  #11 opened 7 years ago by timgluz
  2

• Refactor similarity measures into own module;

  #13 opened 7 years ago by timgluz
  1

• Add Naive Bayes based classifier

  #14 opened 7 years ago by timgluz
  0

• Bug: building index from the `data` folder takes forever;

  #10 opened 7 years ago by timgluz
  1

• Bug: Document terms vector doesnt match with query vector;

  #9 opened 7 years ago by timgluz
  0

• Add crossvalidation for measuring accuracy of models and make them comparable;

  #8 opened 7 years ago by timgluz
  1

• add license matcher based text similarity

  #5 opened 7 years ago by timgluz
  1

• implement core collections and functions

  #6 opened 7 years ago by timgluz
  2

• Add persistence for built indexes to avoid rebuilding it everytime;

  #7 opened 7 years ago by timgluz
  4

• implement reverse index to keep doc id and IDF values

  #3 opened 7 years ago by timgluz
  1

• add term indexer

  #4 opened 7 years ago by timgluz
  2

• collect initial training and test data;

  #1 opened 7 years ago by timgluz
  2