v0.8.1

TextAnalysis v0.8.1

Diff since v0.7.5

Merged pull requests:

• allow DocumentMetadata to hold arbirtary data (#158) (@tanmaykm)
• Directional coom (#264) (@atantos)
• Fixed UNICODE processing with the strip_non_letters flag in src/preprocessing.jl (#265) (@sigmundv)
• ROUGE: fixed sentences calculation and some minor refactoring (#272) (@rssdev10)
• CI: updated scripts. Minimal Julia is 1.6 now (#275) (@rssdev10)
• Code refactoring (#276) (@rssdev10)
• documentation update (#277) (@rssdev10)
• CompatHelper: add new compat entry for Statistics at version 1, (keep existing compat) (#278) (@github-actions[bot])
• Fix/showprogress (#281) (@rssdev10)
• Fix/style improvement (#282) (@rssdev10)

Closed issues:

• error on LDA Julia 0.4 (#37)
• remove_corrupt_utf8() not working (#41)
• remove_corrupt_utf8! giving "no method matching zero" error (#68)
• stemming issue for certain words e.g. providing -> provid (#69)
• rouge_n not defined (#193)
• error strip_spares_terms not defined (#212)
• Eval can be replaced by getfield in tag_scheme! (#242)
• Seems there are some typos in documents (#249)
• StringIndexError when trying to create a StringDocument based on a UTF8 string (#255)
• Converting Corpus to Dataframe not working. (#279)

v0.7.5

TextAnalysis v0.7.5

Diff since v0.7.4

Merged pull requests:

• CompatHelper: add new compat entry for DelimitedFiles at version 1, (keep existing compat) (#269) (@github-actions[bot])
• Clean README, docs and docstrings (#270) (@pitmonticone)
• Update coom.jl (#271) (@ms10596)
• added BLEU score (#273) (@rssdev10)
• Update README.md (#274) (@ms10596)

Closed issues:

• Implementation of cosine similarity? (#215)
• Dependence on BinaryProvider.jl prevents TextAnalysis from working on arm64-apple-darwin natively. (#260)

v0.7.4

TextAnalysis v0.7.4

Diff since v0.7.3

Closed issues:

• PerceptronTagger is not defined (#262)
• Libstemmer not defined for ARM (M1 Mac) (#263)

Merged pull requests:

• Update README.md (#254) (@dunefox)
• Move some docs to TextModels (#256) (@AdarshKumar712)
• fix string indexing in summary (#257) (@ericphanson)
• CompatHelper: bump compat for StatsBase to 0.34, (keep existing compat) (#268) (@github-actions[bot])

v0.7.3

TextAnalysis v0.7.3

Diff since v0.7.2

Closed issues:

• CI is failing on the latest Julia master (#252)

Merged pull requests:

• add cosine similarity calculation (#248) (@hhaensel)
• Latent Dirichlet allocation: display a progress bar during Gibbs sampling (#250) (@DilumAluthge)
• remove write_sub (#253) (@aviks)

v0.7.2

TextAnalysis v0.7.2

Diff since v0.7.1

Closed issues:

• Methods to merge two DocumentTermMatrix instances (#243)

Merged pull requests:

• CompatHelper: bump compat for "DataFrames" to "0.22" (#239) (@github-actions[bot])
• Use Tables.jl, remove explicit DataFrames dependency (#240) (@aviks)
• methods to help manipulate and update DocumentTermMatrix incrementally (#244) (@tanmaykm)
• optimize document term sparse matrix operations (#245) (@tanmaykm)
• fix Project.toml, add Tables compat entry (#246) (@tanmaykm)

v0.7.1

TextAnalysis v0.7.1

Diff since v0.7.0

Closed issues:

• Move models to TextModels.jl (#111)
• Tag a new release (#177)
• Provide libstemmer through Yggdrasil (#204)
• Julia TextAnalysis NERTagger (#214)
• Unable to convert corpus to DataFrame (#236)

Merged pull requests:

• Fix conversion to DataFrame (#237) (@aviks)
• fix link to the docs in README.md (#238) (@gxyd)

v0.7.0

TextAnalysis v0.7.0

Diff since v0.6.0

Closed issues:

• Feature Request: Part of speech tagging (#2)
• Implement Named Entity Recognition (NER) (#117)
• Can a new release be tagged? (#139)
• Need API documentation (#146)
• Extend Naive Bayes Classifier to support the various document types (#152)
• Summarize function throws error for docs with less than 5 sentences. (#153)
• UndefVarError when prepare! called on Corpus (#171)
• Need to export Flux, Tracker (#178)
• Docs and docstring for Sentiment Analysis model needs fixing (#182)
• NaiveBayesClassifier scope error. (#192)
• APIs to avoid datatype constraint between CorpusLoaders.jl and TextAnalysis.jl (#195)
• Add entry for ULMFiT in docs/make.jl (#196)
• Unexpected behaviour of ngram(sd, 3) (#202)
• "resulting" bug (#205)
• Statistical tokenization algorithms (#207)
• Trying to use NaiveBayesClassifier results in UndefVarError (#216)

Merged pull requests:

• Simple document classifier (AKA spam filter) (#106) (@MikeInnes)
• Average Perceptron POS Tagger (Issue #2) (#131) (@ComputerMaestro)
• Remove HTML style tags in preprocessing (#137) (@phereford)
• PR: To address performance issues with stopword removal (#141) (@asbisen)
• Indentation fix patch (#142) (@Ayushk4)
• Fix deprecated function in extended example (#144) (@ViralBShah)
• Add characters to list of punctuations (#145) (@asbisen)
• Add API documentation (#147) (@aquatiko)
• Update ngramizer.jl (#148) (@djokester)
• Add offline Documentation (Docstrings) to the codebase (#150) (@Ayushk4)
• Documentation for Bayes.jl (#151) (@Ayushk4)
• Update summarizer.jl (#154) (@Ayushk4)
• Fix deprecations in show.jl (#155) (@Ayushk4)
• Added ROUGE Score to TextAnalysis.jl (#156) (@djokester)
• allow multiple ngram complexity in NGramDocument, ngrams and ngrammize (#157) (@tanmaykm)
• Update the documentation reflecting changes in show.jl (#159) (@Ayushk4)
• Add functions for Tagging Schemes and Conversion. (#161) (@Ayushk4)
• Conditional Random Fields (#162) (@Ayushk4)
• BM25, Co-occurrence Matrix, faster ROUGE, Fixing LSA. (#165) (@Ayushk4)
• Use datadeps for AvgPerceptronTagger, add pos tagging over document types (#166) (@Ayushk4)
• Named Entity Recognition (#167) (@Ayushk4)
• Add API for Part of Speech Tagging (#169) (@Ayushk4)
• Add favicon to the docs (#170) (@Ayushk4)
• Fix prepare! on strip_whitespace (#172) (@Ayushk4)
• Readme updated. Docs edited to provide API Reference online. (#173) (@Ayushk4)
• ULMFiT (#179) (@aviks)
• Fix Sequence Labelling Models, fixes #178 (#180) (@Ayushk4)
• Drop support for 0.7 and add support for 1.3 (#181) (@Ayushk4)
• Minor fix of doc and docstring of Sentiment Analysis (#184) (@tejasvaidhyadev)
• Remove duplicate entries in Project.toml, and fix a broken build (#189) (@DilumAluthge)
• Bump version number from "0.6.0" to "0.7.0" (#190) (@DilumAluthge)
• Install TagBot as a GitHub Action (#194) (@JuliaTagBot)
• updated docs/make.jl (#198) (@tejasvaidhyadev)
• make DTM type generic (#199) (@baggepinnen)
• bug fix in get_sentiment function (#206) (@tejasvaidhyadev)
• Language Model Interface (#210) (@tejasvaidhyadev)
• Modify loop in initial assignments of lda to use sparse structure. (#213) (@jmoralez)
• export NaiveBayesClassifier (#217) (@agarie)
• Extend NaiveBayesClassifier to support Documents as input #152 (#219) (@KimBue)
• Minor Fixes (#220) (@tejasvaidhyadev)
• LM doc fix (#233) (@tejasvaidhyadev)
• Split project, separate TextModels (#234) (@aviks)

Various fixes

Merge pull request #137 from phereford/master

Remove HTML style tags in preprocessing

v0.5.0 Julia 1.0 Compatibility

Merge pull request #97 from JuliaText/as/towards07

Prepare for 1.0

v0.4.2 Bugfixes

bug fixes for DTM constructor and for remove_patterns (#94)

* Fixes case where DocumentTermMatrix(crps, lex) would construct a dtm of wrong dimensions if a term provided in lex does not occur in the crps

* added DocumentTermMatrix constructor that takes a crps and a prespecified terms vector

* fixed remove_patterns to use nextind() to find starting position of next unstripped substring. Closes #92

* removed leftover info() statement in prepare! test