{"id":19054768,"url":"https://github.com/loginchik/review-sentiment-nlp","last_synced_at":"2026-04-17T06:03:11.713Z","repository":{"id":243579491,"uuid":"812396366","full_name":"loginchik/Review-Sentiment-NLP","owner":"loginchik","description":"ML analysis and model construction based on kinopoisk.ru animation movies reviews","archived":false,"fork":false,"pushed_at":"2024-06-09T22:10:42.000Z","size":48337,"stargazers_count":0,"open_issues_count":1,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-11-12T06:03:26.630Z","etag":null,"topics":["kernel-svm-classifier","kinopoisk","random-forest-classifier","sklearn"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":false,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/loginchik.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,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-06-08T19:12:22.000Z","updated_at":"2024-06-09T22:12:30.000Z","dependencies_parsed_at":"2024-06-10T00:39:37.857Z","dependency_job_id":null,"html_url":"https://github.com/loginchik/Review-Sentiment-NLP","commit_stats":null,"previous_names":["loginchik/review-sentiment-nlp"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/loginchik/Review-Sentiment-NLP","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/loginchik%2FReview-Sentiment-NLP","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/loginchik%2FReview-Sentiment-NLP/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/loginchik%2FReview-Sentiment-NLP/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/loginchik%2FReview-Sentiment-NLP/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/loginchik","download_url":"https://codeload.github.com/loginchik/Review-Sentiment-NLP/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/loginchik%2FReview-Sentiment-NLP/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31917372,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-16T18:22:33.417Z","status":"online","status_checked_at":"2026-04-17T02:00:06.879Z","response_time":62,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":["kernel-svm-classifier","kinopoisk","random-forest-classifier","sklearn"],"created_at":"2024-11-08T23:39:44.702Z","updated_at":"2026-04-17T06:03:11.695Z","avatar_url":"https://github.com/loginchik.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Модель классификации отзывов на мультфильмы\n\nСкрипт для автоматизированного сбора отзывов с Кинопоиска: [Kinopoisk Reviews](https://github.com/loginchik/Kinopoisk-Reviews)\n\nСистема отзывов (рецензий) на Кинопоиске устроена таким образом, что каждый зарегистрированный на сайте пользователь \nможет написать рецензию на фильм и опубликовать её в публику. Перед публикацией пользователю необходимо определить \nсентимент рецензии: позитивная, нейтральная или негативная. В рамках проекта строится предсказательная модель, которая\nна основе 5860 отзывов на различные мультфильмы предсказывает с точностью 70+% сентимент отзыва на основе его текста. \n\nВ изначальных данных наблюдался значительный перекос в сторону позитивных отзывов, поэтому итоговая модель работает \nс двумя классами: позитивный и нейтрально-негативный. Кроме того, в процессе обучения дисбаланс классов компенсировался \nметодом ADASYN. \n\nПрактическое применение модели может быть таким: фильтрация рецензий для анализа плюсов и недостатков анимационных \nработ других режиссёров и совершенствования собственной другими авторами. \n\n## Содержимое проекта\n\n+ Построение модели: [model_construction.ipynb](model_construction.ipynb)\n+ Применение модели к реальным данным: [usage.ipynb](usage.ipynb)\n+ Модель случайного леса (классификатор): [random_forest.joblib](pretrained_models/random_forest.joblib) (precision = 73%)\n+ Модель ядерного метода опорных векторов: [svc.joblib](pretrained_models/svc.joblib) (precision = 79%)\n+ Модель, объединяющая две других модели по принципу sort-voting: [voting_classifier.joblib](pretrained_models/voting_classifier.joblib) (precision = 79%)\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Floginchik%2Freview-sentiment-nlp","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Floginchik%2Freview-sentiment-nlp","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Floginchik%2Freview-sentiment-nlp/lists"}