Build intelligent apps. Master the abstract of machine learning in four hands-on courses.
Suggested by: Coursera (What is Coursera?)
No prior knowledge required
No unnecessary risks
This internship from leading researchers at the University of Washington introduces you to the exciting and in-demand field of machine learning. Through a series of practical cases, you will gain hands-on experience in key areas of machine learning, including:
You will learn how to analyze large and complex data sets, create systems that adapt and improve over time, and build intelligent applications that can make predictions from data.
Learners will implement and apply machine learning prediction, classification, clustering, and information retrieval algorithms on real-world datasets in each course in the specialization. They will emerge with hands-on experience in machine learning and Python programming.
In this course, you will experience machine learning through a series of practical cases. By the end of the first course, you will learn to predict apartment prices based on apartment features, analyze sentiment from user feedback, extract relevant documents, recommend products, and search for images. Through hands-on practice in these cases, you will be able to apply machine learning methods to a wide range of domains.
This course treats machine learning methodology as a black box. Using this abstraction, you will focus on understanding relevant tasks, adapting these tasks to machine learning tools, and evaluating the quality of the output. In subsequent courses, you will explore the components of this black box by examining models and algorithms. Together, these components form the machine learning pipeline, which you will leverage to develop intelligent applications.
In our first test case, predicting apartment prices, you will create models that predict a continuous value (price) from input attributes (area, number of rooms and baths, …). This is just one of many examples where hydration can be applied. Other applications range from predicting health outcomes in medicine, to stock prices in finance, to analyzing the effects of genetic expression.
In this course, you will explore regular linear hydration models for prediction and feature selection tasks. You will be able to deal with very large feature sets and choose between models with different levels of complexity. You will also analyze the impact of different aspects of your data – such as outliers – on the models and predictions you have chosen. To fit these models, you will implement optimization algorithms that can scale to large data sets.
In our Sentiment Analysis test case, you will create models that predict classes (positive/negative sentiments) from input features (review content, user profile information,…). In the second test case of this course, Loan Default Prediction, you will handle financial data and determine when a loan may be risky or safe for the bank. These tasks are examples of classification, one of the most widely used areas of machine learning, with a wide range of applications, including advertising targeting, spam detection, medical diagnostics, and image classification.
In this course, you will build classifiers that deliver top-notch performance across a variety of tasks. You will learn the most successful, widely used techniques in the field, including logistic regression, decision trees, and bounces. You will also design and implement the underlying algorithms that can learn these models at scale, using stochastic gradient descent. You will apply these techniques to real-world, large-scale machine learning tasks. You will also tackle significant tasks you will encounter in real-world ML applications, including handling missing data and measuring accuracy and retrieval to evaluate a classifier. This course is hands-on, action-packed, and includes simulations and demonstrations of how these techniques would behave on real data. We have also included optional content in each module, covering advanced topics for those who want to go even deeper!
A reader is interested in a particular news article and you want to find similar articles to recommend. What is the correct concept of similarity? Moreover, what happens if there are millions of other documents? Every time you want to retrieve a new document, will you have to search through all the other documents? How do you group similar documents together? How can you discover new and emerging topics that the documents are addressing?
In this third case study, Finding Similar Documents, you will examine algorithms based on similarity for retrieval. In this course, you will also examine structured representations for describing documents in a corpus, including clustering and mixed membership models, such as Latent Dirichlet Assignment (LDA). You will implement Expected Optimization (EM) to learn the clusters of documents, and you will have an example of extending the methods using MapReduce.
