Showing posts from February, 2021

Mise en scène

Aditi lent me her workspace today, it's making me happy. The pandemic encouraged her to transform our living room into something very flexible and modular. It's so her, so lush with greenery, part yoga studio, part editorial office, part nonprofit organizing and fundraising space, and ultimately a living space. In this age of mise-en-screen in virtual background, it offers a lot of possibilities. Not pictured is Aditi's "doors" photography series, featuring New Orleans (French quarter) and Montreal (plateau), and our fiction/craft bookshelves.  I've also been playing with the use of cinema to transform TVs from eyesores to art. Jacques Tati's work is particularly suited for it. The mise-en-scène is immaculate. Every frame a perfect tableau. Astonishing sound design: Not music, but musical. Non-narrative: you can absorb the movies over many partial viewings. On a good television, with good audio, the unfolding of a Tati movie is like great painting. It can

Coherence in Machine-Assisted Qualitative Research

This is the first  in a  series of posts  exploring  the notion of coherence when using topic  modeling  methods like Latent Dirichlet Allocation (LDA) for qualitative analysis of texts. I'm trying to think about how mechanically measured coherence matters when you are trying to interpret something qualitatively from a large corpus of texts (with the assistance of machine-learning tools). Having estimated many, many LDA models of  similar  corpora of texts over the last few months for one reason or the other , I have started to observe some methodologically salient patterns in coherence . So  this post will amount to me trying to reflect on the tacit knowledge gained from "doing" topic modeling repeatedly. I hope there is value in this for others attempting to use methods like LDA to support qualitative analysis of texts. To illustrate my reflections, I will refer to a "vignette," comprised of R code, results, and visualizations of LDA analysis, written for p