Other posts in this series: Bank Marketing - Part 1 - Making the call - An Exploratory Data Analysis Bank Marketing - Part 2 - Who will say yes? - Building Classification Models In Part 1, we found that the subscription rate changed a lot between 2008 and 2010, as interest rates fell and the bank changed its campaigns. In Part 2, we built models on a random sample of the calls and tested them on the other calls from the same period. They did well. But when we trained them on the earlier calls and tested them on the later calls, they did much worse.
Other posts in this series: Bank Marketing - Part 1 - Making the call - An Exploratory Data Analysis Bank Marketing - Part 3 - Keeping up with the times - Sliding and Expanding Window Models In Part 1, we explored the Bank Marketing dataset: 41,188 calls in which a Portuguese bank asked its clients to open a term deposit. Only about 1 in 9 clients said yes.
In this post, we build models that predict whether a client will subscribe. A model is a formula that a computer learns from examples. A classification model does not just answer yes or no: it gives each client a probability of saying yes. A bank can use these probabilities to call the most promising clients first.
Other posts in this series: Bank Marketing - Part 2 - Who will say yes? - Building Classification Models Bank Marketing - Part 3 - Keeping up with the times - Sliding and Expanding Window Models A Portuguese bank ran a series of telemarketing campaigns between May 2008 and November 2010. Its staff called clients and asked them to open a term deposit, a savings account that pays a fixed interest rate for a fixed period. The Bank Marketing dataset (Moro, Cortez and Rita, 2014) records 41,188 of these calls: who the client was, how and when they were called, the state of the economy at the time, and whether the client said yes.
Other posts in this series: Ames Housing - Part 1 - Laying the foundation - Building a Data Pipeline Ames Housing - Part 2 - Walkthrough - An Exploratory Data Analysis In Part 1, we built a data pipeline that cleans the Ames Housing dataset, and in Part 2, we explored the finalized dataset.
In this post, we build linear and non-linear models, and we find out how well they predict the SalePrice of properties. A model is a formula that a computer learns from examples. A linear model adds up the features, each multiplied by a number (a coefficient) that it learns from the data. A non-linear model can also learn curved relationships.
Other posts in this series: Ames Housing - Part 1 - Laying the foundation - Building a Data Pipeline Ames Housing - Part 3 - Closing the deal - Building Models In Part 1, we built a data pipeline that cleans the Ames Housing dataset and creates new features. In this post, we explore the finalized dataset. An exploratory data analysis (EDA) is the first study of a dataset before we build models. We look at the features one at a time, then two and three at a time, to find which features affect the sale price. In Part 3, we use what we learn to build models.
Other posts in this series: Ames Housing - Part 2 - Walkthrough - An Exploratory Data Analysis Ames Housing - Part 3 - Closing the deal - Building Models In this case study, we use the Ames Housing dataset to predict the sale prices of houses. The case study has three posts:
In this post, we build a data pipeline: a series of steps that fixes errors in the data, fills in missing values, creates new features and drops features that we do not need. At the end, we save the finalized dataset. In Part 2, we explore the finalized dataset to find which features affect the sale price. In Part 3, we build models that predict the prices. A model is a formula that a computer learns from examples. Neighborhoods of Ames Location is very important for the price of a house. The version of the dataset that we use (from Kaggle) does not have the location of each property. But the full dataset by Dean De Cock, which is in the R package modeldata, has the latitude and longitude of 2930 properties. We use it to show where the neighborhoods are.
Other posts in this series: Diamonds - Part 1 - In the rough - An Exploratory Data Analysis Diamonds - Part 2 - A cut above - Building Linear Models In the previous posts, we found out which attributes of a diamond are important for its price. An exploratory data analysis showed that carat, clarity and color are the most important features to predict price. Then we built linear models to predict the prices, and we chose the best model by its metrics.
Other posts in this series: Diamonds - Part 1 - In the rough - An Exploratory Data Analysis Diamonds - Part 3 - A polished gem - Building Non-linear Models In a previous post in this series, we explored the diamonds dataset. We found that price has a strong correlation with carat, x, y and z, and that log(price) has an almost straight-line relationship with log(carat). For diamonds of similar weight, clarity and color also seemed to help predict the price.
Other posts in this series: Diamonds - Part 2 - A cut above - Building Linear Models Diamonds - Part 3 - A polished gem - Building Non-linear Models In this case study, we explore the diamonds dataset. Then, in the next posts, we build models that predict the price of a diamond. A model is a formula that a computer learns from examples. Some models are linear (they use straight-line relationships) and some are non-linear (they can also learn curves).