Simon Hengchen

Hi! I am a postdoctoral researcher at the University of Helsinki, where I have joined the COMHIS group. I research semantic change in multilingual, unstructured, OCRed, historical textual data and specialise in topic modelling (LDA) with a special interest on the relation between words and ideas, and how they evolve through time.

I was a visiting researcher at the Long Room Hub of Trinity College Dublin for three months In 2015, and for two weeks at the Alan Turing Institute in 2017. In 2018 I have returned twice to the Turing, both times for a bit more than a week.

I have obtained my PhD from the Université libre de Bruxelles.

I speak French, Dutch and English. I can decipher very basic Norwegian Bokmål. I code in Python and dabble in R.

In the Spring 2019, I will be teaching “LDA-H506, Method course in digital humanities II: Intro to NLP for DH (Python)”, an elective course in the Master’s Programme “Linguistic Diversity in the Digital Age”.
One day I will make a better webpage for myself, today is not this day.

My most recent publications are here: tuhat
Others are here: difusion
There's also google scholar: scholar

Most important publications/things are, as of today (2018-09-11):

Hengchen, S., O’Connor, A., Munnelly, G., and Edmond, J., 2016a. Comparing topic model stability across language and size. In Proceedings of the Japanese Association for Digital Humanities Conference 2016.
Hengchen, S., Coeckelbergs, M., van Hooland, S., Verborgh, R., and Steiner, T., 2016b. Exploring archives with probabilistic models: Topic modelling for the valorisation of digitised archives of the European Commission. In First Workshop «Computational Archival Science: digital records in the age of big data», Washington, volume 8.
Hengchen, S., 2017a. When does it mean? Detecting semantic change in historical texts. In Digital Humanities at Oxford Summer School – Text to Tech.
Hengchen, S., 2017b. Detecting Semantic Change Using LDA in Historical Texts: a Case Study on Dutch. In Language Technology Lab Seminars, Cambridge.
De Wilde, M., & Hengchen, S., 2017. Semantic Enrichment of a Multilingual Archive with Linked Open Data. Digital Humanities Quarterly, 11(4).
Hengchen, S., Kanner, A., Marjanen, P., and Mäkelä, E., 2018. Comparing Topic Model Stability Between Finnish, Swedish, English and French. Digital Humanities in the Nordic Countries 2018.
Lathi, L., Vaara, V., Marjanen, J., Roivainen, H., Ijaz, A., Säily, T., Kanner, A., Hill, M., Mäkelä, E., Tolonen, M., 2018. Quantitative analysis of public discourse in Europe 1470-1910. DHBenelux 2018.

Also this book: van Hooland, S., Gillet, F., Hengchen, S., and De Wilde, M., 2016. Introduction aux humanités numériques: méthodes et pratiques. De Boeck supérieur.

Soon a collective work will be available, with a contribution from McGillivray, B., Buning, R., and Hengchen, S. Title of the collective work is "Reassembling the Republic of Letters: Systems, Standards, Scholarship", edited by Howard Hotson and Thomas Wallnig. Title of our contribution is "Extracting topics over time in the Hartlib Papers."