机器学习基础：案例研究（华盛顿大学）

在本课程中，您将从一系列实用的案例研究中获得有关机器学习的动手经验。 在第一门课程的最后，您将研究如何基于房屋特征预测房价，从用户评论中分析情绪，检索感兴趣的文档，推荐产品以及搜索图像。 通过使用这些用例的动手实践，您将能够在广泛的领域中应用机器学习方法。

• 课时1：welcome-to-this-course-and-specialization (42秒)

• 课时2：who-we-are (5分43秒)

• 课时3：machine-learning-is-changing-the-world (3分8秒)

• 课时4：why-a-case-study-approach (1分16秒)

• 课时5：specialization-overview (1分39秒)

• 课时6：how-we-got-into-ml (3分23秒)

• 课时7：who-is-this-specialization-for (4分41秒)

• 课时8：what-you-ll-be-able-to-do (57秒)

• 课时9：the-capstone-and-an-example-intelligent-application (6分31秒)

• 课时10：the-future-of-intelligent-applications (2分19秒)

• 课时11：starting-an-ipython-notebook (5分30秒)

• 课时12：creating-variables-in-python (7分15秒)

• 课时13：conditional-statements-and-loops-in-python (8分8秒)

• 课时14：creating-functions-and-lambdas-in-python (3分31秒)

• 课时15：starting-graphlab-create-loading-an-sframe (4分32秒)

• 课时16：canvas-for-data-visualization (4分9秒)

• 课时17：interacting-with-columns-of-an-sframe (4分29秒)

• 课时18：using-apply-for-data-transformation (5分17秒)

• 课时19：predicting-house-prices-a-case-study-in-regression (1分22秒)

• 课时20：what-is-the-goal-and-how-might-you-naively-address-it (3分47秒)

• 课时21：linear-regression-a-model-based-approach (5分34秒)

• 课时22：adding-higher-order-effects (4分11秒)

• 课时23：evaluating-overfitting-via-training-test-split (6分19秒)

• 课时24：training-test-curves (4分22秒)

• 课时25：adding-other-features (1分14秒)

• 课时26：other-regression-examples (3分28秒)

• 课时27：regression-ml-block-diagram (3分59秒)

• 课时28：loading-exploring-house-sale-data (7分11秒)

• 课时29：splitting-the-data-into-training-and-test-sets (2分34秒)

• 课时30：learning-a-simple-regression-model-to-predict-house-prices-from-house-size (3分54秒)

• 课时31：evaluating-error-rmse-of-the-simple-model (2分29秒)

• 课时32：visualizing-predictions-of-simple-model-with-matplotlib (4分52秒)

• 课时33：inspecting-the-model-coefficients-learned (1分18秒)

• 课时34：exploring-other-features-of-the-data (6分24秒)

• 课时35：learning-a-model-to-predict-house-prices-from-more-features (3分23秒)

• 课时36：applying-learned-models-to-predict-price-of-an-average-house (5分7秒)

• 课时37：applying-learned-models-to-predict-price-of-two-fancy-houses (7分20秒)

• 课时38：analyzing-the-sentiment-of-reviews-a-case-study-in-classification (38秒)

• 课时39：what-is-an-intelligent-restaurant-review-system (4分23秒)

• 课时40：examples-of-classification-tasks (4分51秒)

• 课时41：linear-classifiers (5分6秒)

• 课时42：decision-boundaries (3分57秒)

• 课时43：training-and-evaluating-a-classifier (4分25秒)

• 课时44：whats-a-good-accuracy (3分10秒)

• 课时45：false-positives-false-negatives-and-confusion-matrices (6分28秒)

• 课时46：learning-curves (5分36秒)

• 课时47：class-probabilities (1分53秒)

• 课时48：classification-ml-block-diagram (3分30秒)

• 课时49：loading-exploring-product-review-data (2分53秒)

• 课时50：creating-the-word-count-vector (2分5秒)

• 课时51：exploring-the-most-popular-product (4分37秒)

• 课时52：defining-which-reviews-have-positive-or-negative-sentiment (4分32秒)

• 课时53：training-a-sentiment-classifier (3分17秒)

• 课时54：evaluating-a-classifier-the-roc-curve (4分24秒)

• 课时55：applying-model-to-find-most-positive-negative-reviews-for-a-product (4分43秒)

• 课时56：exploring-the-most-positive-negative-aspects-of-a-product (4分41秒)

• 课时57：document-retrieval-a-case-study-in-clustering-and-measuring-similarity (35秒)

• 课时58：what-is-the-document-retrieval-task (1分31秒)

• 课时59：word-count-representation-for-measuring-similarity (6分57秒)

• 课时60：prioritizing-important-words-with-tf-idf (3分42秒)

• 课时61：calculating-tf-idf-vectors (5分29秒)

• 课时62：retrieving-similar-documents-using-nearest-neighbor-search (2分23秒)

• 课时63：clustering-documents-task-overview (2分28秒)

• 课时64：clustering-documents-an-unsupervised-learning-task (4分39秒)

• 课时65：k-means-a-clustering-algorithm (4分0秒)

• 课时66：other-examples-of-clustering (6分2秒)

• 课时67：clustering-and-similarity-ml-block-diagram (7分1秒)

• 课时68：loading-exploring-wikipedia-data (5分21秒)

• 课时69：exploring-word-counts (5分53秒)

• 课时70：computing-exploring-tf-idfs (7分9秒)

• 课时71：computing-distances-between-wikipedia-articles (5分38秒)

• 课时72：building-exploring-a-nearest-neighbors-model-for-wikipedia-articles (3分18秒)

• 课时73：examples-of-document-retrieval-in-action (4分16秒)

• 课时74：recommender-systems-overview (41秒)

• 课时75：where-we-see-recommender-systems-in-action (19秒)

• 课时76：building-a-recommender-system-via-classification (4分3秒)

• 课时77：collaborative-filtering-people-who-bought-this-also-bought (6分15秒)

• 课时78：effect-of-popular-items (1分5秒)

• 课时79：normalizing-co-occurrence-matrices-and-leveraging-purchase-histories (6分10秒)

• 课时80：the-matrix-completion-task (5分18秒)

• 课时81：recommendations-from-known-user-item-features (6分0秒)

• 课时82：predictions-in-matrix-form (3分45秒)

• 课时83：discovering-hidden-structure-by-matrix-factorization (7分32秒)

• 课时84：bringing-it-all-together-featurized-matrix-factorization (30秒)

• 课时85：a-performance-metric-for-recommender-systems (5分30秒)

• 课时86：optimal-recommenders (2分10秒)

• 课时87：precision-recall-curves (7分12秒)

• 课时88：recommender-systems-ml-block-diagram (4分53秒)

• 课时89：loading-and-exploring-song-data (5分43秒)

• 课时90：creating-evaluating-a-popularity-based-song-recommender (5分8秒)

• 课时91：creating-evaluating-a-personalized-song-recommender (5分53秒)

• 课时92：searching-for-images-a-case-study-in-deep-learning (23秒)

• 课时93：what-is-a-visual-product-recommender (3分52秒)

• 课时94：using-precision-recall-to-compare-recommender-models (4分6秒)

• 课时95：application-of-deep-learning-to-computer-vision (5分41秒)

• 课时96：deep-learning-performance (3分5秒)

• 课时97：demo-of-deep-learning-model-on-imagenet-data (2分57秒)

• 课时98：other-examples-of-deep-learning-in-computer-vision (1分30秒)

• 课时99：challenges-of-deep-learning (2分22秒)

• 课时100：deep-features (6分44秒)

• 课时101：deep-learning-ml-block-diagram (3分11秒)

• 课时102：loading-image-data (3分53秒)

• 课时103：training-evaluating-a-classifier-using-raw-image-pixels (6分14秒)

• 课时104：training-evaluating-a-classifier-using-deep-features (8分9秒)

• 课时105：loading-image-data (2分54秒)

• 课时106：creating-a-nearest-neighbors-model-for-image-retrieval (1分56秒)

• 课时107：querying-the-nearest-neighbors-model-to-retrieve-images (5分42秒)

• 课时108：querying-for-the-most-similar-images-for-car-image (1分53秒)

• 课时109：displaying-other-example-image-retrievals-with-a-python-lambda (4分2秒)

• 课时110：you-ve-made-it (40秒)

• 课时111：deploying-an-ml-service (3分13秒)

• 课时112：what-happens-after-deployment (7分5秒)

• 课时113：open-challenges-in-ml (8分40秒)

• 课时114：where-is-ml-going (6分29秒)

• 课时115：whats-ahead-in-the-specialization (5分38秒)

• 课时116：thank-you (1分32秒)