**Knowledge discovery and semantic learning in the framework of axiomatic fuzzy set theory**

Axiomatic fuzzy set (AFS) theory facilitates a way on how to transform data into fuzzy sets (membership functions) and implement their fuzzy logic operations, which provides a flexible and powerful tool for representing human knowledge and emulate human recognition process. In recent years, AFS theory has received increasing interest. In this survey, we report the current developments of theoretical research and practical advances in the AFS theory. We first review some notion and foundations of the theory with an illustrative example, then, we focus on the various extensions of AFS theory for knowledge discovery, including clustering, classification, rough sets, formal concept analysis, and other learning tasks. Due to its unique characteristics of semantic representation, AFS theory has been applied in multiple domains, such as business intelligence, computer vision, financial analysis, and clinical data analysis. This survey provides a comprehensive view of these advances in AFS theory and its potential perspectives.

**The Difference Between Managing Large and Small Data Science Teams**

As advanced analytics and data science have matured into must-have skills, data science groups within large companies have themselves become much larger. This has led to some unique problems and solutions that you´ll want to consider as your own DS group grows larger.

**Batch normalization in Neural Networks**

We normalize the input layer by adjusting and scaling the activations. For example, when we have features from 0 to 1 and some from 1 to 1000, we should normalize them to speed up learning. If the input layer is benefiting from it, why not do the same thing also for the values in the hidden layers, that are changing all the time, and get 10 times or more improvement in the training speed.

**Testing machine learning interpretability techniques**

The importance of testing your tools, using multiple tools, and seeking consistency across various interpretability techniques.

**Models are about what changes, and what doesn’t**

Approaching the model build by thinking about how the sale of ice cream will change, as the temperature changes, and considering what to keep constant has led me down the road of differential equations. With a small data set like this, starting from a simple model and reasoning any additional complexity has helped me to develop a better understanding of what a reasonable underlying data generating process might be. Indeed, generating data from the model before using the data, helped to check my understanding of how the model might perform.

**Running Python inside the RStudio Server IDE**

A great many R users will have to run some python code from time to time, and this short video from our Head of Data Engineering, Mark Sellors outlines one approach you can take that doesn’t mean leaving the RStudio IDE.

**Deep Learning for Time Series Forecasting: Predicting Sunspot Frequency with Keras**

In this post we will examine making time series predictions using the sunspots dataset that ships with base R. Sunspots are dark spots on the sun, associated with lower temperature. Here´s an image from NASA showing the solar phenomenon.

Advertisements