BOT OR NOT

Code and data for:
Li, Z., Jiang, Q., Wu, Z., Liu, A., Wu, H., Huang, M., Huang, K. & Ku, Y. (in press). Towards human-compatible autonomous car: A study of non-verbal Turing test in automated driving with affective transition modelling. IEEE Transactions on Affective Computing. https://doi.org/10.1109/TAFFC.2023.3279311

---

Outreach

• A poster for Proceedings of the 45th Annual Conference of the Cognitive Science Society is available here.
• A poster for the Social & Affective Neuroscience Society (SANS) Annual Meeting 2023 is available on ResearchGate.
• A 4.2-minute video for the SANS 2023 is available on Twitter. The related slides are available on ResearchGate.
• The slides for 2022 National Doctoral Forum on Brain-Computer Intelligence and Psychology are available here.
• A poster for the 3rd Macau Symposium on Cognitive and Brain Sciences is available on ResearchGate.
• The slides for International Graduate Forum on Language Cognitive Science are available here.
• The slides for the 1st International Symposium on Addiction and Decision Making are available here.
• The slides for Greater Bay Area Young Scholar Forum on Psychological Science are available here.
• Social media: Twitter, WeChat (in Chinese), LinkedIn, Mastodon.
• arXiv preprint.

---

Structure

root
 ├── bert4keras                    # Adapted from https://github.com/bojone/bert4keras
 ├── data                          # Processed affective transition data & regression data & original behavioural data
 │    ├── av_data 
 │    ├── olra_data
 │    └── xls_data
 ├── data_prep.py                  # To provide functions for behavioural data processing
 ├── at_prep.py                    # To provide functions for affective transition generation
 ├── at_generator.py               # To generate affective transition
 ├── sdt.py                        # To provide functions for model building and nested leave-one-out cross-validation
 ├── evaluate.py                   # To provide functions for model evaluation
 ├── fig&tbl.ipynb                 # To plot figures 2-8 and appendix figure 1 and make tables 2 and appendix table 1
 ├── tbl1_SDT-AT_Original.ipynb    # Results of SDT-AT (Original) models
 ├── tbl1_SDT-AT_PLM-tf.ipynb      # Results of SDT-AT (PLM-tf) models
 ├── tbl1_SDT-AT_PLM-wv.ipynb      # Results of SDT-AT (PLM-wv) models
 ├── tbl1_Baselines                # Results of Baselines models
 │    ├──  ml_baselines
 │    └──  naive_baselines.ipynb
 ├── appx_tbl1_olra.Rmd            # Results for appendix table 1
 ├── teaser_image.png
 ├── requirements.txt
 ├── LICENSE
 └── README.md

Note 1: to properly run all scripts, you may need to set the root of this repository as your working directory and install Python modules and packages listed in requirements.txt which I used in my MacBook Pro (M1 Max).
Note 2: to generate affective transition, you may need to download related pre-trained models or weights listed in at_generator.py.

---

Citation

@article{li2023Bot,
  author = {Li, Zhaoning and Jiang, Qiaoli and Wu, Zhengming and Liu, Anqi and Wu, Haiyan and Huang, Miner and Huang, Kai and Ku, Yixuan},
  title = {Towards human-compatible autonomous car: A study of non-verbal Turing test in automated driving with affective transition modelling}
  journal = {IEEE Transactions on Affective Computing}, 
  doi = {https://doi.org/10.1109/TAFFC.2023.3279311}
  year = {2023},
}

---

For bug reports, please contact Zhaoning Li (yc17319@umac.mo, or @lizhn7).

LICENSE

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, which gives you the right to re-use and adapt, as long as you note any changes you made, and provide a link to the original source.