Predicting sales values using some past data (multiple features) by machine learning

[Abdessamad7]

Hello Community,
I have a dataset that containes those features : Year, Retailer country (country name), product type (string), Performence(float), sales (float $)
I have Year from 2004 to 2007 in this dataset
I want to use a machine learning algorithm in Knime so as to predict sales in 2008 for each couple (country,product type)

Any help thanks in advance

[HansS]

Hi @Abdessamad7

I think predicting sales or predicting whatever is absolutly possible with KNIME. Take a look at the KNIME Hub, and search for regression model workflow see the examples and get inspired. Or go to the KNIME learning hub. And also with KNIME comes the EXAMPLES server with lots of workflows dealing with predictive modelling.
gr. Hans

[mlauber71]

Here we have a few examples for regression models with KNIME:

Regression models (numeric Target)

Predictive analytics for Sales

I can think of a few examples that might show you the possible directions This site has an introduction to model building with KNIME and uses an example that could be similar to your question: Concerning the predictions you want you will have to think about how to set up the analysis and prediction. Would you make a prediction for every single product and dealer and what role does the time play in your analysis. Sales within a month or a week. Then the quality of a model will very much de…

Forecasting sales per customer for the next 360 days

@ScottF provided us with simple workflow that shows how you would set up a regression model and do an evaluation using the RMSE ( Root-mean-square deviation) => the lower the better. The KNIME numeric scorer node has you covered there. You could switch out the type of model used: We had a discussion about that here: And also further information about ways to determine the quality of a model (mostly 1/0 but check out the link). I have to see if I have an example with a regression targe…

Evaluate a Linear Regression model

Typically you would use the Scorer nodes. Then there are a few entries dealing with numeric targets, you might want to check them. Regression models (numeric Target) predict how many future visitors a restaurant will receive (with H2O.ai)

How to identify the Top 100 features selected from MLP model

I was thinking about feeding the prediction of the model back to a correlation node (linear in this case) and then use the absolute value of the correlation to determine the strongest correlations towards the predicted Target. The data is from a Kaggle competition (https://www.kaggle.com/c/bnp-paribas-cardif-claims-management) with some transformations and only 500 random cases. So the numbers are just there to demonstrate the nodes. The calculation can be quite costly. [31] kn_example_mlp_m…

predict how many future visitors a restaurant will receive (with H2O.ai)

Predict Demand Combining the Power of KNIME & H2O | KNIME