RUB Research School

Detail

  1. You are here:
  2. Startseite
  3. About
  4. Map of International Research Activities (MIRA)
  5. Detail
Back

Author
Name Nickel, Robert
Research field Natural Language Processing and Speech Signal Processing
Career stage professor
Home university/institution Bucknell University
Department/Research unit at home university/institution Electrical and Computer Engineering
Chair/Working group at home institution Cognitive Signal Processing

International activity
Country Germany
Location Bochum
University Ruhr-Universität Bochum (RUB)
Fund Research School VIP
Type of activity research stay
Period starts 22-07-2021
ends 12-08-2021
Keywords Natural Language Processing, Authorship Verification, Misinformation Detection
Report The ongoing collaboration between Prof. Nickel, doctoral candidate Benedikt Bönninghoff, and Prof. Kolossa from the Cognitive Signal Processing Research Group at the Ruhr-Universität Bochum is centered around the development of theoretical concepts and technical procedures for an "intelligent" automatic extraction of information from text documents (see [1-7]) and multimedia sources (see [8-15]). An example is provided by the advances of the group in forensic text comparison. Forensic text comparison tries to answer the question if two texts with unknown authorship were written by the same authors or not. It has traditionally been done by linguists who employ discipline-specific models to categorize a text. Deep machine learning mechanisms can be trained to accomplish the same task with very little bias towards predefined categories. Results of the work, so far, have been published in multiple articles at international venues and journals [1-7]. The work has also led to first places in two international research challenges, the PAN Authorship Verification Challenges [A1, A2], as well as the award of a research grant from the BMBF (Bundesministerium für Bildung und Forschung) under a Forschung-Agil Call [G1] with the title "KI-unterstütztes Assistenzsystem für die Crowdsourcing-basierte Erkennung von über digitale Plattformen verbreiteter Desinformation".

With support from the RUB Research School PLUS VIP Program it was possible for Prof. Nickel to visit the Ruhr-Universität Bochum multiple times over the past five years, first from July 10th until August 4th, 2017, then from February 5th until February 9th, 2018, and followed by May 29th until August 3rd, 2018. In 2019 he also visited twice, from May 27th until June 12th and from July 22th until August 16th, followed by a short visit in early 2020 from January 20th until January 27th. Additional visits in 2020 had to be postponed due to the COVID crisis and resumed in 2021 from July 23rd until Aug 13th. All of these visits were devoted to in-person work with Benedikt Bönninghoff, Prof. Kolossa, and her research group. Prof. Nickel is co-advisor for Benedikt Bönninghoff who is slated to graduate with his doctorate in mid 2022.

Joint Publications:

[1] "Self-Calibrating Neural-Probabilistic Model for Authorship Verification Under Covariate Shift," by B. Boenninghoff, D. Kolossa, and R. M. Nickel, 12th International Conference of the CLEF Association, Bucharest, Romania, September 21-24, 2021.

[2] "O2D2: Out-Of-Distribution Detector to Capture Undecidable Trials in Authorship Verification," by B. Boenninghoff, R. M. Nickel, and D. Kolossa, Notebook for PAN 2021 at the Conference and Labs of the Evaluation Forum (CLEF), Bucharest, Romania, September 21-24, 2021.

[3] "Variational Autoencoder with Embedded Student-t Mixture Model for Authorship Attribution," by B. Boenninghoff, S. Zeiler, R. M. Nickel, and D. Kolossa, May 28, 2020, arXiv:2005.13930 [cs.LG], 28th International Conference on Computational Linguistics (COLING), Barcelona, Spain, December 8-13, 2020.

[4] "Deep Bayes Factor Scoring for Authorship Verification," by B. Boenninghoff, J. Rupp, R. M. Nickel, and D. Kolossa, Notebook for PAN 2020 at the Conference and Labs of the Evaluation Forum (CLEF), Thessaloniki, Greece, September 22-25, 2020.

[5] "Autorschaftsanalyse - Verstellungsstrategien und Möglichkeiten der automatisierten Erkennung," by B. Bönninghoff, S. Hessler, D. Kolossa, K. Kucharczik, R. M. Nickel, and K. Pittner, Datenschutz und Datensicherheit - DuD, Springer Verlag, vol. 43, pp. 691–699, Nov 2019.

[6] "Explainable Authorship Verification in Social Media via Attention-based Similarity Learning," B. Boenninghoff, S. Hessler, D. Kolossa and R. M. Nickel, IEEE International Conference on Big Data (Big Data), Los Angeles, CA, USA, December 9-12, 2019, pp. 36-45.

[7] "Similarity Learning for Authorship Verification in Social Media," by B. Boenninghoff, R. M. Nickel, S. Zeiler, and D. Kolossa, Proceedings of the International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Brighton, UK, May 12-17, 2019, pp. 2457-2461.

[8] "Improving Audio-Visual Speech Recognition Using Deep Neural Networks with Dynamic Stream Reliability Estimates," by H. Meutzner, N. Ma, R. M. Nickel, C. Schymura, and D. Kolossa, Proceedings of the International Conference on Acoustics, Speech, and Signal Processing (ICASSP), New Orleans, Louisiana, March 5-9, 2017, pp. 5320-5324.

[9] "Unsupervised Classification of Voiced Speech and Pitch Tracking Using Forward-Backward Kalman Filtering," by B. T. Bönninghoff, R. M. Nickel, S. Zeiler, and D. Kolossa, Proceedings of the 12th Speech Communication Conference of the Information Technology Society of the Association of German Engineers (ITG-VDE Fachtagung Sprachkommunikation), Paderborn, Germany, October 5-7, 2016.

[10] "Dynamic Stream Weighting for Turbo-Decoding-Based Audiovisual ASR," by S. Gergen, S. Zeiler, A. H. Abdelaziz, R. M. Nickel, D. Kolossa, Proceedings of the 17th Annual Conference of the International Speech Communication Association (INTERSPEECH), San Francisco, California, September 8-12, 2016, pp. 2135-2139.

[11] "Robust Audiovisual Speech Recognition Using Noise-Adaptive Linear Discriminant Analysis," by S. Zeiler, R. M. Nickel, N. Ma, G. J. Brown, and D. Kolossa, Proceedings of the International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Shanghai, China, March 20-25, 2016, pp. 2797-2801.

[12] "Corpus-Based Speech Enhancement with Uncertainty Modeling and Cepstral Smoothing," R. M. Nickel, R. F. Astudillo, D. Kolossa, and R. Martin, IEEE Transactions on Audio, Speech, and Language Processing, Vol. 21, No. 5, May 2013, pp. 983-997.

[15] "Inventory-Based Audio-Visual Speech Enhancement," D. Kolossa, R. M. Nickel, S. Zeiler, and R. Martin, 13th Annual Conference of the International Speech Communication Association (INTERSPEECH), Portland, Oregon, September 9-13, 2012.

International Research Challenge Awards:

[A1] June 2021: The implementation of an algorithm developed by Benedikt Boenninghoff, Robert M. Nickel, and Dorothea Kolossa won the 2021 PAN Authorship Verification Challenge. The team with name “boenninghoff21” placed first in the large dataset challenge against more than 10 international research groups. The competition results are published at:
https://pan.webis.de/clef21/pan21-web/author-identification.html

[A2] July 2020: The implementation of an algorithm developed by Benedikt Boenninghoff, Julian Rupp, Robert M. Nickel, and Dorothea Kolossa won the 2020 PAN Authorship Verification Challenge. Our team with name “boenninghoff20” placed first in both the large dataset and the small dataset challenge against 10 international research groups. The competition results are published at:
https://pan.webis.de/clef20/pan20-web/author-identification.html

Awarded Grant Proposal:

[G1] Since December 2021: A project with the title "KI-unterstütztes Assistenzsystem für die Crowdsourcing-basierte Erkennung von über digitale Plattformen verbreiteter Desinformation" (Artificial Intelligence driven Support System for Crowd-Sourcing based Detection of Misinformation on Digital Platforms) is currently funded by the BMBF (Bundesministerium für Bildung und Forschung) under a Forschung-Agil Call. The proposal was jointly written by Prof. Nickel (Bucknell University), Prof. Kolossa (RUB), Benedikt Boenninghoff (RUB), Prof. Scheffler (RUB), Publisher David Schraven (Head of the German Fact Checking Organization CORRECTIV), and Prof. Gostomzyk (Technical University Dortmund). The lead PI on the proposal is Prof. Kolossa.
Back