A new framework ties word vectors and classifiers for better language modeling.
problem Inefficiencies in learning and parameter usage in conventional language models.
method Introduces a novel loss framework that ties input embeddings and output projections, reducing parameters.
result State-of-the-art performance on Penn Treebank with various network models.
SWESA learns word embeddings with document labels for sentiment analysis.
problem Sentiment analysis using limited text data.
method SWESA uses supervised learning to optimize word embeddings and classifier performance.
result SWESA outperforms existing methods in sentiment analysis.
Simple model classifies traffic tweets from Twitter using word embeddings.
problem Sparsity and curse of dimensionality in bag-of-words representation for traffic tweets.
method Proposes a word embedding-based framework to classify traffic-related tweets from non-traffic ones.
result State-of-the-art test accuracy of 95.9% achieved with a simple model.
This paper classifies tweets into positive and negative sentiments using distributed word and sentence representations.
problem Classifying tweets into positive and negative sentiments.
method Used distributed representations of words and sentences, and LSTM and CNN networks for classification.
result Achieved accuracies as high as 81%.
BERT improves Chinese word segmentation performance.
problem Chinese word segmentation task.
method Applying BERT to CWS task using benchmark datasets.
result BERT can improve performance even with inconsistent labels.
Improved aspect detection from few seed keywords.
problem Fine-grained aspect detection from user reviews is labor-intensive.
method Weakly supervised co-training with student-teacher approach.
result Significant improvement in F1 scores over previous methods.
The study counts curves on a once-punctured torus with self-intersections.
problem Counting closed curves with self-intersections on a once-punctured torus.
method Combinatorial classification of curves with given word-length and self-intersections.
result Determination of curve counts with zero, one, and arbitrary self-intersections.
Estimating the difficulty level of math word problems is an important task for many educational applications. Identification of relevant and irrelevant sentences in math word problems is an important step for calculating the difficulty levels of such problems. This paper addresses a novel application of text categoriza…
The study evaluates 62 classifiers for detecting online toxic comments.
problem Identifying and classifying toxic online commentary.
method Systematic evaluation of 62 classifiers representing 19 algorithmic families on the Jigsaw dataset.
result Simple bad word lists are most predictive of offensive commentary.
The paper uses geometry to assess how hard examples are for NLP models.
problem Challenges in NLP datasets and classifiers, especially with shallow features.
method Information geometry to quantify example difficulty, exploring BERT, CNN, and fasttext.
result Deep learning models are vulnerable to word substitutions in difficult examples.
Word embeddings improve classification of research stages.
problem Classifying research stages using conventional methods.
method Used pre-trained and custom word embeddings for classification.
result Custom embeddings outperform general embeddings for research classification.
WME generates document embeddings from word embeddings, outperforming state-of-the-art techniques.
problem Lack of unsupervised document embeddings from pre-trained word embeddings.
method Word Mover's Embedding (WME) approach.
result WME consistently matches or outperforms state-of-the-art techniques on various text classification and similarity tasks.
This paper improves NLP interpretability by using sentence segments instead of words.
problem Limitations of word-based sampling in explaining complex BERT models.
method Using sentence segments as elementary building blocks for NLP interpretability.
result Improved fidelity of the explainer on a benchmark classification task.
Enhanced neural model detects misinformation by considering author context.
problem Plateaued accuracy in traditional rumor classifiers.
method Ensemble neural architecture with word-level attention and author context.
result Promising results on Twitter dataset, detecting rumors more accurately.
Anchors explain text model decisions by highlighting key words.
problem Explaining neural network decisions for text data.
method Formalized Anchors for text classification, analyzing different models.
result Neural network words with highest partial derivatives are selected by Anchors.
Automatically classifying the tissues types of Region of Interest (ROI) in medical imaging has been an important application in Computer-Aided Diagnosis (CAD), such as classification of breast parenchymal tissue in the mammogram, classify lung disease patterns in High-Resolution Computed Tomography (HRCT) etc. Recently…
Classifies complex hyperbolic triangle groups by types.
problem Classifying complex hyperbolic triangle groups.
method By types defined by the ellipticity of two short words.
result Improves Schwartz conjecture.
This paper reviews different word embeddings for sentiment classification using deep learning.
problem Handling large textual data with simple ML algorithms.
method Word embedding strategies implemented on an Amazon Review Dataset.
result Different word embeddings improve accuracy in sentiment classification.
Machine learning explains text document categorization decisions.
problem Understanding how text documents are categorized by machine learning models.
method Layer-wise relevance propagation (LRP) to trace predictions back to individual words.
result Word-based ML models can be made more comprehensible through LRP.
Classifies flat knots up to 8 crossings using Lyndon words.
problem Classifying flat knots up to a certain number of crossings.
method Using matchings on Lyndon words and various knot invariants.
result Distinguished all flat knots up to 7 crossings except for five pairs.
A single BLSTM network tackles ambiguous words in text data.
problem Ambiguity in text data, especially in technical domains.
method Proposes a single Bidirectional LSTM network for all ambiguous words.
result Comparable performance to top WSD algorithms on SensEval-3 benchmark.
We construct the infinite sequence of invariants for curves in surfaces by using word theory that V. Turaev introduced. For plane closed curves, we add some extra terms, e.g. the rotation number. From these modified invariants, we get the Arnold's basic invariants and some other invariants. We also express how these in…
New tool detects weak and strong Islamophobic hate speech on social media.
problem Detecting Islamophobic hate speech on social media is challenging due to its varied nature.
method Built a multi-class classifier distinguishing between non-Islamophobic, weak Islamophobic, and strong Islamophobic content using GloVe word embeddings.
result Accuracy of 77.6% and balanced accuracy of 83% on a dataset of 109,488 tweets.
Layer-wise relevance propagation (LRP) is a recently proposed technique for explaining predictions of complex non-linear classifiers in terms of input variables. In this paper, we apply LRP for the first time to natural language processing (NLP). More precisely, we use it to explain the predictions of a convolutional n…
Cross-language learning allows us to use training data from one language to build models for a different language. Many approaches to bilingual learning require that we have word-level alignment of sentences from parallel corpora. In this work we explore the use of autoencoder-based methods for cross-language learning …
Study examines biases in clinical word embeddings, revealing performance gaps across groups.
problem Biases in clinical word embeddings leading to performance differences across groups.
method Pretrained BERT models on MIMIC-III, fill-in-the-blank method, fairness evaluation on clinical tasks.
result Classifiers trained from BERT representations exhibit statistically significant differences in performance across groups.
Amobee wins WASSA 2018 emotion prediction with language models and LSTM.
problem Predicting emotions from tweets without explicit mentions.
method Ensemble system of language models and LSTM with CNN attention.
result 1st place with macro F1 score of 0.7145.
Few-shot NER classification with limited data.
problem Limited annotated data for Named Entity Recognition.
method Prototypical Network + Transfer Learning
result Well-performing classifiers trained on only 20 instances.
Proposes a new feature-based evaluation method for explaining Deep Learning models in text classification.
problem Lack of consideration for linguistic dependencies in existing attribution-based explanations.
method Investigates perturbations based on embedded features removal from intermediate layers of Convolutional Neural Networks.
result Visualization tool assists analysts in understanding model predictions better.
This paper justifies and improves lexicon-based classification without labeled data.
problem Lack of justification for lexicon-based classification and its lower accuracy compared to supervised methods.
method Derives probabilistic justification and learns weights from co-occurrence statistics.
result Lexicon-based classification can be improved without labeled data, offering higher accuracy.
Proposes RTL model for sentiment classification and key word detection in online reviews.
problem Sentiment classification and key word detection in online reviews for hospitality industry.
method Regularized Text Logistic (RTL) regression model.
result RTL model achieves satisfactory classification performance and identifies key word features.
Improved TSC with BOSS and SP techniques.
problem Comparing BOP and BOSS for time series classification.
method Deconstructed and measured components of BOP and BOSS, adapted CV techniques.
result SP with BOSS significantly more accurate than benchmarks.
Adversarial consistency depends on the uniqueness of adversarial Bayes classifiers.
problem Consistency of adversarial surrogate losses is not guaranteed.
method Connected consistency of adversarial surrogate losses to the uniqueness of adversarial Bayes classifiers.
result A convex surrogate loss is statistically consistent for adversarial learning if and only if the adversarial Bayes classifier is unique.
IITK wins FinSim 2020 task on financial hypernym detection.
problem Classifying financial terms into hypernym concepts in an external ontology.
method Used context-dependent and context-independent word embeddings (Word2vec and BERT) for classification.
result Ranked 1st based on mean rank and accuracy metrics.
Anchors explains text classifiers by highlighting key words.
problem Interpreting machine learning models, especially for text classifiers.
method Formalizes Anchors' algorithm and analyzes its behavior on linear text classifiers.
result Anchors produces meaningful results on linear text classifiers.
Paper classifies Brazilian music genres using song lyrics with BLSTM network.
problem Classifying Brazilian music genres from lyrics.
method Used BLSTM network combined with SVM, Random Forest, and word embeddings.
result BLSTM outperforms other models with an F1-score of 0.48.
Paper automates substation signal matching using machine learning.
problem Manual matching of customer data with substation signal names.
method Bagged token classifier that votes on signal names based on customer data.
result The method outperforms standard classifiers in accuracy and efficiency.
Classifies CAD model descriptions and names from product websites.
problem Distinguishing product descriptions from other text and identifying product names.
method Paragraph vectors, character-level LSTM, word embeddings LSTM tagger.
result Promising results for distinguishing product descriptions and names.
We explore two techniques which use color to make sense of statistical text models. One method uses in-text annotations to illustrate a model's view of particular tokens in particular documents. Another uses a high-level, "words-as-pixels" graphic to display an entire corpus. Together, these methods offer both zoomed-i…
Hybrid QC system for Bengali questions using smart data balancing.
problem Classifying factoid questions in Bengali.
method Two-stage approach with 1D CNN for coarse classification and SGD for fine classification.
result Effectiveness of smart data balancing technique in improving classification accuracy.
Improved stock prediction using news features and RNN.
problem Predicting stock prices with high accuracy.
method Extracted news features, optimized seed words, calculated positive polar, constructed news features, proposed RNN model.
result Our method improves stock prediction accuracy by over 5%.
This thesis focuses on gaining linguistic insights into textual discussions on a word level. It was of special interest to distinguish messages that constructively contribute to a discussion from those that are detrimental to them. Thereby, we wanted to determine whether "I"- and "You"-messages are indicators for eithe…
Neural nets classify Thai Lukthung songs from other genres.
problem Classifying Thai Lukthung songs for personalized music recommendations.
method Developed neural networks using lyrics and audio features.
result Combined model yields F1 score of 0.86 for Lukthung classification. Identifying the type of font (e.g., Roman, Blackletter) used in historical documents can help optical character recognition (OCR) systems produce more accurate text transcriptions. Towards this end, we present an active-learning strategy that can significantly reduce the number of labeled samples needed to train a font…
Framework improves text classification under budget constraints.
problem Building robust text classifiers with limited computational resources.
method Jointly trains a selector to identify relevant words and passes them to a classifier, with a data aggregation scheme.
result Improves classifier performance and speeds up model with minimal accuracy loss.
New findings on algebraic structure of hyperbolic graph braid groups.
problem Classifying and understanding the algebraic structure of hyperbolic graph braid groups.
method Analyzing specific graph types (sun and pulsar graphs) and proving theorems about their braid groups.
result 3-strand braid groups of sun graphs are free, while most pulsar graphs contain surface subgroups.
Paper tackles toxic comment detection using deep learning.
problem Automatic detection of toxic comments on the internet.
method Designs binary classification and regression-based approaches using DNN.
result BERT fine-tuning outperforms other methods.
The paper presents a method to preserve privacy in text analysis using calibrated noise.
problem Accurately learning from user data while maintaining privacy.
method Calibrated multivariate perturbations applied to word embeddings to achieve geo-indistinguishability.
result The method provides better privacy guarantees than baseline models with minimal utility loss.