NeuroXL Package

Those who are involved into data mining and use Microsoft Office Excel as their main tool, might struggle if circumstances require cluster analysis or prediction of indicators such as product demand. Fortunately, there are software solutions for such issues and one of them is NeuroXL Package. This Microsoft Excel add-in will offer users the ability to perform clustering analysis and estimation, through its two main components, NeuroXL Clusterizer and NeuroXL Predictor, from the same developer.

Being easy to deploy and featuring no additional configuration steps, the add-in will be readily available in users’ installation of Microsoft Office Excel, under the Add-ins tab. Its menu contains its two main components, NeuroXL Clusterizer and NeuroXL Predictor, which fortunately, users can access simultaneously.

The cluster analysis can be performed using the three-tab array that allows users to easily indicate the cell range where the clustering will be deployed, together with the parameters for the neural network. Subsequently, the prediction process can be initiated by providing the cell range for the sample training and the corresponding estimation parameters.

Users will be able to analyze their spreadsheet data in detail and the add-in will employ some of the most intriguing architecture for its algorithm: human neural networks. The parameters that control the distribution of clusters and the estimation can be easily set up and do not require any expert knowledge.

However, having said that, the analysis process might turn into a somewhat empirical approach, especially since no preview of the changes occurring when adjusting the parameters is provided. This might not be an important issue, but it might have been beneficial if the application also offered a form of preview for the parameter adjustments, at least through simple examples.

This add-in could prove to be a good companion for those who require tools for achieving an advanced analysis of their Excel spreadsheet data. It will allow users to either assess clustering in their data or perform predictions, with customizable parameters for training samples and activation functions. Based  on the architecture of human neural networks, the add-in’s algorithm brings together the capabilities of NeuroXL Clusterizer and NeuroXL Predictor.