Towards a comprehensive visualization of structure in data

Dimensional data reduction methods are fundamental to explore and visualize large data sets. Basic requirements for unsupervised data exploration are simplicity, flexibility and scalability. However, current methods show complex parameterizations and strong computational limitations when exploring large data structures across scales. Here, we focus on the t-SNE algorithm and show that a simplified parameter setup with a single control parameter, namely the perplexity, can effectively balance local and global data structure visualization. We also designed a chunk\&mix protocol to efficiently parallelize t-SNE and explore data structure across a much wide range of scales than currently available. Our parallel version of the BH-tSNE, namely pt-SNE, converges to good global embedding, comparable to state-of-the-art solutions, though the chunk\&mix protocol adds little noise and decreases the accuracy at the local scale. Nonetheless, we show that simple post-processing can efficiently restore local scale visualization, without any loss of precision at the global scales. We expect the same approach to apply to faster embedding algorithms other than BH-tSNE, like FIt-SNE or UMAP, thus, extending the state-of-the-art and leading to more comprehensive data structure visualization and analysis.

bigMap: Big Data Mapping with Parallelized t-SNE

We introduce an improved unsupervised clustering protocol specially suited for large-scale structured data. The protocol follows three steps: a dimensionality reduction of the data, a density estimation over the low dimensional representation of the data, and a final segmentation of the density landscape. For the dimensionality reduction step we introduce a parallelized implementation of the well-known t-Stochastic Neighbouring Embedding (t-SNE) algorithm that significantly alleviates some inherent limitations, while improving its suitability for large datasets. We also introduce a new adaptive Kernel Density Estimation particularly coupled with the t-SNE framework in order to get accurate density estimates out of the embedded data, and a variant of the rainfalling watershed algorithm to identify clusters within the density landscape. The whole mapping protocol is wrapped in the bigMap R package, together with visualization and analysis tools to ease the qualitative and quantitative assessment of the clustering.