Google’s latest is an Add-on that uses machine learning to make predictions in Google Sheets.
Dubbed Smart Autofill, the add-on uses the already available spreadsheet data to learn patterns and predicts the missing entries. It is unlike the standard “Autofill” feature of Sheets, which is capable of filling in simple values like calendar dates, days of the week, ordered numbers, etc.
Prediction API is Google’s cloud based machine learning service, utilizing which, the add-on trains several linear as well as non-linear classification and regression models, explains Google’s Research Blog. The model with the smallest misclassification error (for categorical data) or root-mean-squared error (for numeric data) which is calculated by using cross-validation on the labeled (non-empty) set of examples, is selected automatically.
“The accuracy of Smart Autofill (as well as the accuracy of the estimated error) depends on many factors, including the amount and quality of the data provided,” the blog says, as not all datasets will be ideally suited for machine learning.
“Much of Google’s work on language, speech, translation, and visual processing relies on machine learning, where we construct and apply learning algorithms that make use of labeled data in order to make predictions for new data,” write Konstantin Davydov, Software Engineer and Afshin Rostamizadeh, Research Scientist at Google.
Leveraging machine learning algorithms to learn patterns in spreadsheet data to build a model and infer unknown values is the essential idea behind this add-on. This tutorial provides an idea of the range of problems where Smart Autofill can be effective.
Read more here.
(Image credit: Google)
