das-boot/scite

Code and data for paper 'Causality extraction based on self-attentive BiLSTM-CRF with transferred embeddings'

Hi there 👋

---

I'm Zhaoning!

• 👣 I am a second-year PhD student working with Prof. Haiyan Wu in Affective, Neuroscience and Decision-making Laboratory (A. N. D Lab) in the psychology program at the University of Macau (Macao, China).

  Before that, I was a research assistant supervised by Prof. Yixuan Ku in Memory and Emotion Laboratory (MELab) at the Department of Psychology, Sun Yat-Sen University (SYSU, Guangzhou, China). Before joining this lab, I worked as an NLP engineer in the Department of Big Data and Artificial Intelligence at Tianpeng Technology Co., Ltd (Guangzhou, China).

  I received a B.E. degree in Information Security in 2016 and an M.E. degree in Software Engineering in 2018, both from SYSU.

• 💡 Enlightened by Heinrich Heine's famous quote, 'Thought precedes action, as lighting does thunder', I devoted myself to theoretical research after familiarising myself with empirical research and application. Specifically, I aspire to make contributions to a 'virtuous circle' in which shared insights into artificial intelligence and social psychology will advance the development of both fields.

• 🌊 Research interests: Social cognition, social neuroscience, mentalising, artificial social intelligence, natural language processing

---

Recent Research Articles

• Towards human-compatible autonomous car: A study of non-verbal Turing test in automated driving with affective transition modelling
  BibTeX

  @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},
  }

• Every individual makes a difference: A trinity derived from linking individual brain morphometry, connectivity and mentalising ability
  BibTeX

  @article{li2023Trinity,
    author = {Zhaoning Li and Qunxi Dong and Bin Hu and Haiyan Wu},
    title = {Every individual makes a difference: A trinity derived from linking individual brain morphometry, connectivity and mentalising ability},
    journal = {Human Brain Mapping},
    volume = {44},
    number = {8},
    pages = {3343-3358},
    doi = {https://doi.org/10.1002/hbm.26285},
    url = {https://onlinelibrary.wiley.com/doi/abs/10.1002/hbm.26285},
    year = {2023},
  }
  

---

scite's Issues

缺少news-backward-0.4.1.pt

你好，在Create Contextual String Embeddings with Flair的步骤中，如下代码
charlm_embedding_backward = FlairEmbeddings('your path to /news-backward-0.4.1.pt')
在我的本地路径中是一个空文件，请问还有哪里可以下载到这个文件呢

Some wrong in tag2triplet module

scite/models/tag2triplet.py

Line 205 in 1e082be

record.append(([np.abs(e[0]-e[1])

For record.append(([np.abs(e[0]-e[1]) for e in i], list(i))), I think the code should be like that: record.append((sum([np.abs(e[0]-e[1]) for e in i]), list(i)))

About data_prep.ipynb

Hi, i;m tring to use data_prep.ipynb to generate the trainning data,but stuck in this cell:Create Word Embeddings with Wiki-Extvec
when i try to do "Copy from extvec weights of words that appear in index2word",following error comes out:
`---------------------------------------------------------------------------
ConnectionRefusedError Traceback (most recent call last)
d:\apps\anaconda\envs\py36\lib\site-packages\urllib3\connection.py in _new_conn(self)
169 conn = connection.create_connection(
--> 170 (self._dns_host, self.port), self.timeout, **extra_kw
171 )

d:\apps\anaconda\envs\py36\lib\site-packages\urllib3\util\connection.py in create_connection(address, timeout, source_address, socket_options)
95 if err is not None:
---> 96 raise err
97

d:\apps\anaconda\envs\py36\lib\site-packages\urllib3\util\connection.py in create_connection(address, timeout, source_address, socket_options)
85 sock.bind(source_address)
---> 86 sock.connect(sa)
87 return sock
`
Did you have meet the same problem?And would you mind tell your requirments for this work?
I download the wiki_extvec from https://public.ukp.informatik.tu-darmstadt.de/reimers/2017_english_embeddings/wiki_extvec.gz;
is this right?

It's my fault,I didn't download corenlp..

please put it on HuggingFace!

Hi there,
this is a great tool why you don't put it on hugging face?
Cheers!

训练集174 1915 3311标注有点小错误

Hello I noticed that you missed the Fig. 4 in your preprint version.

I am can wait to see your amazing works!!!!

some Problem in Chain CRF.py

def _fetch_mask(self):
mask = None

    if self._inbound_nodes:
        mask = self._inbound_nodes[0].input_masks[0]

    return mask

I haven't found the corresponding function for the input_masks here. Is there anything that can be replaced?

add_weight in CRF has issues with multiple names

Running the code "as-is" I see

/content/ChainCRF.py in build(self, input_shape)
302 name='U',
303 regularizer=self.U_regularizer,
--> 304 constraint=self.U_constraint)
305
306 self.b_start = self.add_weight((n_classes, ),

TypeError: add_weight() got multiple values for argument 'name'

Is the code maintained?

missing documents

“w_filePath = 'data/index/index_w.pkl'"
Excuse me, How can I get the missing documents?

希望能了解到拓展语料集的标注规范: )

论文中提到了文中使用的语料是对 semeval 2010 task 8 的拓展，希望能了解到你们标注的 准则是怎样的。感谢！

您好，你的代码提供的的数据不全，可以提供一下吗，index_w.pkl，extvec_embedding.npy谢谢

您好，你的代码提供的的数据不全，可以提供一下吗，index_w.pkl，extvec_embedding.npy谢谢
class Data:
def init(self, args):
self.word2index, self.index2word = pickle.load(
open(args.file_path+'index/index_w.pkl', 'rb'))
self.embedding = np.load(
open(args.file_path+'embedding/extvec_embedding.npy', 'rb'))

Model Log links

The README.md lists two links for the model logs, a google drive link and a Baidu Netlink. Do they point to the same model log; if I download the mode log from the google drive link do I still need to download the model log from the Baidu link?