Brain Computer Interface (BCI), especially systems for recognizing brain signals using deep learning after characterizing these signals as EEG (Electroencephalography), is one of the important research topics that arouse the interest of many research ers currently. Convolutional Neural Nets (CNN) is one of the most important deep learning classifiers used in this recognition process, but the parameters of this classifier have not yet been precisely defined so that it gives the highest recognition rate and the lowest possible training and recognition time. This research proposes a system for recognizing EEG signals using the CNN network, while studying the effect of changing the parameters of this network on the recognition rate, training time, and recognition time of brain signals, as a result the proposed recognition system was achieved 76.38 % recognition rate, And the reduction of classifier training time (3 seconds) by using Common Spatial Pattern (CSP) in the preprocessing of IV2b dataset, and a recognition rate of 76.533% was reached by adding a layer to the proposed classifier.
Classification of hierarchical text using geometric deep learning: the case of clinical trials corpus
We consider the hierarchical representation of documents as graphs and use geometric deep learning to classify them into different categories. While graph neural networks can efficiently handle the variable structure of hierarchical documents using t he permutation invariant message passing operations, we show that we can gain extra performance improvements using our proposed selective graph pooling operation that arises from the fact that some parts of the hierarchy are invariable across different documents. We applied our model to classify clinical trial (CT) protocols into completed and terminated categories. We use bag-of-words based, as well as pre-trained transformer-based embeddings to featurize the graph nodes, achieving f1-scoresaround 0.85 on a publicly available large scale CT registry of around 360K protocols. We further demonstrate how the selective pooling can add insights into the CT termination status prediction. We make the source code and dataset splits accessible.
Dialogue summarization comes with its own peculiar challenges as opposed to news or scientific articles summarization. In this work, we explore four different challenges of the task: handling and differentiating parts of the dialogue belonging to mul tiple speakers, negation understanding, reasoning about the situation, and informal language understanding. Using a pretrained sequence-to-sequence language model, we explore speaker name substitution, negation scope highlighting, multi-task learning with relevant tasks, and pretraining on in-domain data. Our experiments show that our proposed techniques indeed improve summarization performance, outperforming strong baselines.
Application of Deep Learning Methods to SNOMED CT Encoding of Clinical Texts: From Data Collection to Extreme Multi-Label Text-Based Classification
Concept normalization of clinical texts to standard medical classifications and ontologies is a task with high importance for healthcare and medical research. We attempt to solve this problem through automatic SNOMED CT encoding, where SNOMED CT is o ne of the most widely used and comprehensive clinical term ontologies. Applying basic Deep Learning models, however, leads to undesirable results due to the unbalanced nature of the data and the extreme number of classes. We propose a classification procedure that features a multiple-step workflow consisting of label clustering, multi-cluster classification, and clusters-to-labels mapping. For multi-cluster classification, BioBERT is fine-tuned over our custom dataset. The clusters-to-labels mapping is carried out by a one-vs-all classifier (SVC) applied to every single cluster. We also present the steps for automatic dataset generation of textual descriptions annotated with SNOMED CT codes based on public data and linked open data. In order to cope with the problem that our dataset is highly unbalanced, some data augmentation methods are applied. The results from the conducted experiments show high accuracy and reliability of our approach for prediction of SNOMED CT codes relevant to a clinical text.
Modelling and understanding dialogues in a conversation depends on identifying the user intent from the given text. Unknown or new intent detection is a critical task, as in a realistic scenario a user intent may frequently change over time and diver t even to an intent previously not encountered. This task of separating the unknown intent samples from known intents one is challenging as the unknown user intent can range from intents similar to the predefined intents to something completely different. Prior research on intent discovery often consider it as a classification task where an unknown intent can belong to a predefined set of known intent classes. In this paper we tackle the problem of detecting a completely unknown intent without any prior hints about the kind of classes belonging to unknown intents. We propose an effective post-processing method using multi-objective optimization to tune an existing neural network based intent classifier and make it capable of detecting unknown intents. We perform experiments using existing state-of-the-art intent classifiers and use our method on top of them for unknown intent detection. Our experiments across different domains and real-world datasets show that our method yields significant improvements compared with the state-of-the-art methods for unknown intent detection.
Existing text style transfer (TST) methods rely on style classifiers to disentangle the text's content and style attributes for text style transfer. While the style classifier plays a critical role in existing TST methods, there is no known investiga tion on its effect on the TST methods. In this paper, we conduct an empirical study on the limitations of the style classifiers used in existing TST methods. We demonstrated that the existing style classifiers cannot learn sentence syntax effectively and ultimately worsen existing TST models' performance. To address this issue, we propose a novel Syntax-Aware Controllable Generation (SACG) model, which includes a syntax-aware style classifier that ensures learned style latent representations effectively capture the sentence structure for TST. Through extensive experiments on two popular text style transfer tasks, we show that our proposed method significantly outperforms twelve state-of-the-art methods. Our case studies have also demonstrated SACG's ability to generate fluent target-style sentences that preserved the original content.
Comparative Analysis of Fine-tuned Deep Learning Language Models for ICD-10 Classification Task for Bulgarian Language
The task of automatic diagnosis encoding into standard medical classifications and ontologies, is of great importance in medicine - both to support the daily tasks of physicians in the preparation and reporting of clinical documentation, and for auto matic processing of clinical reports. In this paper we investigate the application and performance of different deep learning transformers for automatic encoding in ICD-10 of clinical texts in Bulgarian. The comparative analysis attempts to find which approach is more efficient to be used for fine-tuning of pretrained BERT family transformer to deal with a specific domain terminology on a rare language as Bulgarian. On the one side are used SlavicBERT and MultiligualBERT, that are pretrained for common vocabulary in Bulgarian, but lack medical terminology. On the other hand in the analysis are used BioBERT, ClinicalBERT, SapBERT, BlueBERT, that are pretrained for medical terminology in English, but lack training for language models in Bulgarian, and more over for vocabulary in Cyrillic. In our research study all BERT models are fine-tuned with additional medical texts in Bulgarian and then applied to the classification task for encoding medical diagnoses in Bulgarian into ICD-10 codes. Big corpora of diagnosis in Bulgarian annotated with ICD-10 codes is used for the classification task. Such an analysis gives a good idea of which of the models would be suitable for tasks of a similar type and domain. The experiments and evaluation results show that both approaches have comparable accuracy.
CS-UM6P at SemEval-2021 Task 1: A Deep Learning Model-based Pre-trained Transformer Encoder for Lexical Complexity
Lexical Complexity Prediction (LCP) involves assigning a difficulty score to a particular word or expression, in a text intended for a target audience. In this paper, we introduce a new deep learning-based system for this challenging task. The propos ed system consists of a deep learning model, based on pre-trained transformer encoder, for word and Multi-Word Expression (MWE) complexity prediction. First, on top of the encoder's contextualized word embedding, our model employs an attention layer on the input context and the complex word or MWE. Then, the attention output is concatenated with the pooled output of the encoder and passed to a regression module. We investigate both single-task and joint training on both Sub-Tasks data using multiple pre-trained transformer-based encoders. The obtained results are very promising and show the effectiveness of fine-tuning pre-trained transformers for LCP task.
hub at SemEval-2021 Task 7: Fusion of ALBERT and Word Frequency Information Detecting and Rating Humor and Offense
This paper introduces the system description of the hub team, which explains the related work and experimental results of our team's participation in SemEval 2021 Task 7: HaHackathon: Detecting and Rating Humor and Offense. We successfully submitted the test set prediction results of the two subtasks in the task. The goal of the task is to perform humor detection, grade evaluation, and offensive evaluation on each English text data in the data set. Tasks can be divided into two types of subtasks. One is a text classification task, and the other is a text regression task. What we need to do is to use our method to detect the humor and offensive information of the sentence as accurately as possible. The methods used in the results submitted by our team are mainly composed of ALBERT, CNN, and Tf-Idf algorithms. The result evaluation indicators submitted by the classification task are F1 score and Accuracy. The result evaluation index of the regression task submission is the RMSE. The final scores of the prediction results of the two subtask test sets submitted by our team are task1a 0.921 (F1), task1a 0.9364 (Accuracy), task1b 0.6288 (RMSE), task1c 0.5333 (F1), task1c 0.0.5591 (Accuracy), and task2 0.5027 (RMSE) respectively.
CS-UM6P at SemEval-2021 Task 7: Deep Multi-Task Learning Model for Detecting and Rating Humor and Offense
Humor detection has become a topic of interest for several research teams, especially those involved in socio-psychological studies, with the aim to detect the humor and the temper of a targeted population (e.g. a community, a city, a country, the em ployees of a given company). Most of the existing studies have formulated the humor detection problem as a binary classification task, whereas it revolves around learning the sense of humor by evaluating its different degrees. In this paper, we propose an end-to-end deep Multi-Task Learning (MTL) model to detect and rate humor and offense. It consists of a pre-trained transformer encoder and task-specific attention layers. The model is trained using MTL uncertainty loss weighting to adaptively combine all sub-tasks objective functions. Our MTL model tackles all sub-tasks of the SemEval-2021 Task-7 in one end-to-end deep learning system and shows very promising results.