{"id":13696367,"url":"https://github.com/mimno/jsLDA","last_synced_at":"2025-05-03T17:31:00.135Z","repository":{"id":8195165,"uuid":"9625263","full_name":"mimno/jsLDA","owner":"mimno","description":"An implementation of latent Dirichlet allocation in javascript","archived":false,"fork":false,"pushed_at":"2022-08-01T21:36:08.000Z","size":1611,"stargazers_count":181,"open_issues_count":5,"forks_count":51,"subscribers_count":19,"default_branch":"master","last_synced_at":"2024-08-03T18:20:56.991Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"JavaScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":"thorstenb/odpdown","license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/mimno.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2013-04-23T14:37:17.000Z","updated_at":"2024-05-23T09:08:18.000Z","dependencies_parsed_at":"2022-08-06T21:16:01.253Z","dependency_job_id":null,"html_url":"https://github.com/mimno/jsLDA","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mimno%2FjsLDA","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mimno%2FjsLDA/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mimno%2FjsLDA/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mimno%2FjsLDA/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/mimno","download_url":"https://codeload.github.com/mimno/jsLDA/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":224369495,"owners_count":17299908,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-08-02T18:00:38.909Z","updated_at":"2024-11-13T00:30:24.779Z","avatar_url":"https://github.com/mimno.png","language":"JavaScript","funding_links":[],"categories":["Models","JavaScript"],"sub_categories":["Latent Dirichlet Allocation (LDA) [:page_facing_up:](https://www.jmlr.org/papers/volume3/blei03a/blei03a.pdf)"],"readme":"jsLDA\n=====\n\nAn implementation of latent Dirichlet allocation in javascript. A [live demonstration](http://mimno.infosci.cornell.edu/jsLDA/jslda.html) is available.\n\nInstructions:\n------------\n\nWhen you first load the page, it will request a file containing documents and a file containing stopwords. The default example is a corpus of paragraphs from US State of the Union speeches.\n\nClick the \"Run 50 iterations\" button to start training. The browser may appear to freeze for a while.\nInitially all words have been assigned randomly to topics.\nWe train a model by cycling through every word token in the documents and sampling a topic for that word.\nAn \"iteration\" corresponds to one pass through the documents.\n\nThe topics on the left side of the page should now look more interesting. Run more iterations if you would like -- there's probably still a lot of room for improvement after only 50 iterations.\n\nOnce you're satisfied with the model, you can click on a topic from the list on the right to sort documents in descending order by their use of that topic. Proportions are weighted so that longer documents will come first.\n\nYou can also explore correlations between topics by clicking the \"Topic Correlations\" tab. This view shows a force directed layout with connections between topics that have correlations above a certain threshold. You can control this threshold with the slider: a low cutoff will display more edges, while a high cutoff will remove all but the strongest correlations.\n\nTopic correlations are actually pointwise mutual information scores. This score measures whether two topics occur\nin the same document more often than we would expect by chance. Previous versions of this script calculated correlations\non logratios; PMI is simpler to calculate.\n\nUsing your own corpus:\n---------------------\n\nTo use your own corpus, the best way is to place the files in this repository in the document root of a web server.\nReplace the files `documents.txt` and `stoplist.txt` with your own corpus and stop list.\nThe documents file is a tab-delimited text file with one document per line. Each line has three fields:\n\n    [doc ID] [tab] [label] [tab] [text...]\n\n(this is the default format for Mallet). The \"label\" field is currently unused, but I plan to support timestamps, labels, etc.\n\nThe format for stopwords is one word per line. The \"Vocabulary\" tab may be useful in customizing a stoplist. Unicode is supported, so most languages that have meaningful whitespace (ie not CJK) should work.\n\nThe page works best in Chrome. Safari and Firefox work too, but may be considerably slower. It doesn't seem to work in IE.\n\nDownload results:\n----------------\n\nYou can create reports about your topic model. Hit the `Downloads` tab.\nReports are in CSV format. The `sampling state` file contains the same information as a Mallet state file, but in a more compact format. \n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmimno%2FjsLDA","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmimno%2FjsLDA","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmimno%2FjsLDA/lists"}