• 1. What is NextBrain?

  NextBrain is a next-generation auto-machine learning tool that allows you to quickly improve your decision-making process using data. By bringing data directly from your spreadsheets or from many other available sources, you can create professional machine learning models in a few clicks. You don't need to code or have advanced statistical knowledge. NextBrain will walk you through each step, making suggestions but leaving it up to you to decide what to do next. NextBrain takes a step forward iFew readers

• 2. Workspaces and models

  The two basic elements of NextBrain are workspaces and models. Workspaces are equivalent to folders where we can store models. We can share workspaces with others, and this will include all models inside. Workspaces allow you to easily share data between models. A machine learning model is built from an input dataset and an algorithm trained with this dataset. So in models, we can have both data and trained algorithms. (https://storage.crisp.chat/users/helpdesk/websiteFew readers

• 3. Input data

  Data is the fuel of Machine Learning. Algorithms are used to uncover patterns from this data so that new predictions can be made or actionable insights can be acquired. NewxtBrain works with labeled tabular data, which is organized in columns and rows, like in a spreadsheet table. Labeled means that each column must have an identifying name. NextBrain accepts data coming from different sources: Directly from Google Sheets Data stored in a hard disk with a spreadsheet format (xls, xlsx, xlsmFew readers

• 4. Create a model

  To build a model, you need two things: a data set with the target feature we want to predict, and an algorithm that learns from this data to predict this target. In NextBrain, the user has to only select the data set and identify the target. NextBrain will prepare this data and select the best algorithm to learn this data.Few readers

• 5. Train a model

  After we've chosen a dataset, we have to select a column (a feature) we want to predict. We will call this column ‘target’. Then we are going to use a supervised learning technique, that is, the algorithm will learn how to relate the dataset with this target. This process is called learning. The term ‘supervised’ comes from the idea that the learning will be conditioned or supervised by this target. The model will then be trained. NextBrain will prepare the data and choose the best algorithmFew readers

• 6. Make predictions

  After we have trained the model, we can make predictions with new data. With 'new data', we mean data different from the data we used to train the model. There are two ways to enter this new data: by entering a row selecting each column value, or with an external file (as described in Input data) containing several rows. This input must contain the same columns that were used to train the model, except for the target column. As a result of the training, we will get a new column with the predicFew readers

• 7. Get insights

  NextBrain will provide you with several options to get insights about a machine learning model and its results, including: The model's performance metrics: Evaluation of the model's performance on a train and test datasets. Metrics such as accuracy, precision, recall, MAE, MAPE, and R2 score can provide an indication of how well the model is performing. Examine the confusion matrix: A confusion matrix can give you more insight into the model's performance, by showing the number of true posFew readers