Summer read: a new paper on model-based clustering just appeared in Computo!

Julien Jacques and Brendan Thomas Murphy publish a new method for clustering multivariate count data. The method combines feature selection and clustering, and is based on conditionally independent Poisson mixture models and Poisson generalized linear models.

On simulations, the Adjusted Rand Index (ARI) of the model with selected variables is close to the optimal ARI obtained with the true clustering variables.

The paper and accompanying R code are available at https://computo-journal.org/published-202507-jacques-count-data/

Exciting news, our paper is out!

"Behavioral Clusters and Lesion Distributions in Ischemic Stroke, Based on NIHSS Similarity Network" on Springer Journal of Healthcare Informatics Research https://rdcu.be/efgma

With my co-first-author Andrea Zanola and co-authors, we explore the relations between behavioral measures of impairment after stroke, and the underlying brain lesions.
Rather than focusing on covariances at the population level, we first cluster individual behavioral phenotypes, and then explore the typical and significant lesions of each cluster.

Our technique, Repeated Spectral Clustering is performed on a similarity network (derived from the General Distance Measure, handy for ordinal scales!), and the partitions are statistically robust thanks to the aggregation of results from multiple random initializations.

We end up with 5 clusters, 3 of which show reknown principal components of deficits (Left Motor, Righ Motor, Language), and their associate lesions.

Interestingly, this multi-item and multimodal approach allows to distinguish different etiologies for the same deficits, thanks to their different behavioral associations, and the different lesions characterizing each cluster. Even when the single NIHSS measure is a bit "vague"...

We hope that popularizing the General Distance Measure, Repeated Spectral Clustering and this clustering perspective aside of PCA / CCA studies can inspire multimodal approaches in other neuroscientific and biomedical domains!

Many thanks to our co-authors, Antonio Luigi Bisogno, Silvia Facchini, Lorenzo Pini, Manfredo Atzori and Maurizio Corbetta for data, analytic and medical insights, and their guidance throughout the whole process!

A Methodology For The Multitemporal Analysis Of Land Cover Changes And Urban Expansion Using Synthetic Aperture Radar (SAR) Imagery - A Case Study Of The Aburrá Valley In Colombia
--
https://doi.org/10.3390/rs17030554 <-- shared paper
--
#GIS #spatial #mapping #SyntheticApertureRadar #SAR #remotesensing #multitemporalanalysis #landcover #landcoverchange #clustering #kurtosis #fuzzylogic #kernelbasedmethod #machinelearning #spatialanalysis #spatiotemporal #geostatistics #model #modeling #AburráValley #Columbia #urban #urbanexpansion #population #growth #topography #monitoring #satellite #sentinel #valley #landuse #distribution #infrastructure #building #roads #naturalresources #environmental #conservation #monitoring #multitemporal

Hi all #Rstats enthusiasts!
I'm looking for someone who has time now to conduct a review of a piece of software for Journal of Open Source Software (JOSS). Details are here:
https://github.com/openjournals/joss-reviews/issues/7319

The review process is quite simple - you get a checklist and you run some tests. It's all open, on GitHub.

How clustering works with localization in Google Search https://www.seroundtable.com/google-search-clustering-localization-38531.html

Google on the difference between clustering and canonicalization: "Clustering is basically taking the pages that we think are the same. And then canonicalization is, from those pages, which one is the best one" @johnmu said https://www.seroundtable.com/google-search-clustering-canonicalization-38529.html

Stanford Researchers Map ‘White-Only’ Properties In Santa Clara Co. Using AI [ historic deeds / covenants ]
--
https://www.kron4.com/news/bay-area/stanford-researchers-map-white-only-properties-in-santa-clara-co-using-ai/ <-- shared media article
--
https://dho.stanford.edu/wp-content/uploads/Covenants.pdf <-- shared research
--
https://reglab.github.io/racialcovenants/static/maps/dotmap_lot_level.html <-- link to shared webmap
--
#GIS #spatial #mapping #California #deeds #property #racial #racism #redlining #covenenants #race #minorities #propertyrecords #discrimination #history #historical #USHistory #legalreform #records #AI #machinelearning #openlargelanguagemodel #model #modeling #geography #clustering #demographics #spatialanalysis #spatiotemporal

in the #arXiv

2D watershed void clustering for probing the cosmic large-scale structure

by Yingxiao Song and co-authors
https://arxiv.org/abs/2410.04898

Rest in Peace old mate. C. Gordon Bell was a pioneer in Computer Scalability. He took the notion of #Clustering from its infancy to productization. All Computer Architects have stood on his shoulders over the past decades. #Scalability #Clusters #DEC.
An #AmericanTreasure in the #Engineering domain. #GordonBell.

https://www.nytimes.com/2024/05/21/technology/c-gordon-bell-dead.html?smid=url-share

@puppygirlhornypost well, AFAICT from people who used #DragonflyBSD (like @fuchsiii ) it's optimized for #Clustering with #HAMMER & #HAMMER2 filesystem as well as #LWKT which do allow higher throughput that scales I/O and network across multi-socket and -threaded architectures...

https://en.wikipedia.org/wiki/DragonFly_BSD

1/n
Our pre-print is finally out!
Here's my first #paperthread
In this work, co-authors and I clustered ischaemic stroke patients profiles, and recovered common patterns of cognitive, sensorimotor damage.

...Historically many focal lesions to specific cortical areas were associated with specific distinction, but most strokes involve subcortical regions and bring multivariate patterns of deficits.
To characterize those patterns, many studies have turned to correlation analysis, factor analysis, PCA, focusing on the relations among variables==domains of impairments...

https://www.medrxiv.org/content/10.1101/2023.11.08.23297808

Hey #library folks ,

do you want to cluster your book editions with the well-known Work-set algorithm from #OCLC, but you don't find a suitable reusable tool?

I recently faced this issue while working on the #BELTRANS project at KBR (Royal Library of Belgium). All I found were many research papers describing the clustering and a few implementations that required me to install 2010-style Java software stacks.

So I decided to write an easily reusable small #Python script that follows the ideas of the Work-set algorithm: clustering based on descriptive keys. Nothing more, nothing less.

Check my blog post for more information and have a look at the script.

blog post: https://doi.org/10.59350/4hd4r-1tk44

script: https://doi.org/10.5281/zenodo.10011416