Loading...
Search for:
labeled-data
0.006 seconds
How will your tweet be received? predicting the sentiment polarity of tweet replies
, Article 15th IEEE International Conference on Semantic Computing, ICSC 2021, 27 January 2021 through 29 January 2021 ; 2021 , Pages 370-373 ; 9781728188997 (ISBN) ; Monajem, M ; Christlein, V ; Heinrich, P ; Nicolaou, A ; Naderi Boldaji, H ; Lotfinia, M ; Evert, S ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2021
Abstract
Twitter sentiment analysis, which often focuses on predicting the polarity of tweets, has attracted increasing attention over the last years, in particular with the rise of deep learning (DL). In this paper, we propose a new task: predicting the predominant sentiment among (first-order) replies to a given tweet. Therefore, we created RETwEET, a large dataset of tweets and replies manually annotated with sentiment labels. As a strong baseline, we propose a two-stage DL-based method: first, we create automatically labeled training data by applying a standard sentiment classifier to tweet replies and aggregating its predictions for each original tweet; our rationale is that individual errors...
Unsupervised cross-subject BCI learning and classification using riemannian geometry
, Article 24th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2016, 27 April 2016 through 29 April 2016 ; 2016 , Pages 307-312 ; 9782875870278 (ISBN) ; Shamsollahi, M. B ; Jutten, C ; Congedo, M ; Sharif University of Technology
i6doc.com publication
2016
Abstract
The inter-subject variability poses a challenge in cross-subject Brain-Computer Interface learning and classification. As a matter of fact, in cross-subject learning not all available subjects may improve the performance on a test subject. In order to address this problem we propose a subject selection algorithm and we investigate the use of this algorithm in the Riemannian geometry classification framework. We demonstrate that this new approach can significantly improve cross-subject learning without the need of any labeled data from test subjects
Online Detection Of Multi-Modal Fake News
, M.Sc. Thesis Sharif University of Technology ; Rabiee, Hamid Reza (Supervisor) ; Fazli, Mohammad Amin (Supervisor)
Abstract
Today, social networks have provided an environment for disseminating all kinds of information and news among the people. One of the challenges of ex-panding social networks is reading and trusting fake news propagating against accurate information. Fake news is false information that the author intention-ally produces and publishes. Due to the destructive effects of spreading fake news, determining the trustworthiness of news is one of the critical issues in the so-cial, political, and economic fields. In this research, we worked on detecting fake news on social media using multimodal data. To solve the problem of fake news detection, we used the text and images of the information. Transfer...
A transfer learning algorithm based on linear regression for between-subject classification of EEG data
, Article 25th International Computer Conference, Computer Society of Iran, CSICC 2020, 1 January 2020 through 2 January 2020 ; 2020 ; Sardouie, S. H ; Foroughmand Aarabi, M. H ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2020
Abstract
Classification is the most important part of brain-computer interface (BCI) systems. Because the neural activities of different individuals are not identical, using the ordinary methods of subject-dependent classification, does not lead to high accuracy in betweensubject classification problems. As a result, in this study, we propose a novel method for classification that performs well in between-subject classification. In the proposed method, at first, the subject-dependent classifiers obtained from the train subjects are applied to the test trials to obtain a set of scores and labels for the trials. Using these scores and the real labels of the labeled test trials, linear regression is...
Nonlinear unsupervised feature learning: How local similarities lead to global coding
, Article Proceedings - 12th IEEE International Conference on Data Mining Workshops, ICDMW 2012 ; 2012 , Pages 506-513 ; 9780769549255 (ISBN) ; Rabiee, H. R ; Tahaei, M. S ; Salavati, E ; Sharif University of Technology
2012
Abstract
This paper introduces a novel coding scheme based on the diffusion map framework. The idea is to run a t-step random walk on the data graph to capture the similarity of a data point to the codebook atoms. By doing this we exploit local similarities extracted from the data structure to obtain a global similarity which takes into account the nonlinear structure of the data. Unlike the locality-based and sparse coding methods, the proposed coding varies smoothly with respect to the underlying manifold. We extend the above transductive approach to an inductive variant which is of great interest for large scale datasets. We also present a method for codebook generation by coarse graining the data...
Supervised neighborhood graph construction for semi-supervised classification
, Article Pattern Recognition ; Volume 45, Issue 4 , April , 2012 , Pages 1363-1372 ; 00313203 (ISSN) ; Rabiee, H. R ; Sharif University of Technology
Abstract
Graph based methods are among the most active and applicable approaches studied in semi-supervised learning. The problem of neighborhood graph construction for these methods is addressed in this paper. Neighborhood graph construction plays a key role in the quality of the classification in graph based methods. Several unsupervised graph construction methods have been proposed that have addressed issues such as data noise, geometrical properties of the underlying manifold and graph hyper-parameters selection. In contrast, in order to adapt the graph construction to the given classification task, many of the recent graph construction methods take advantage of the data labels. However, these...
Unilateral semi-supervised learning of extended hidden vector state for Persian language understanding
, Article NLP-KE 2011 - Proceedings of the 7th International Conference on Natural Language Processing and Knowledge Engineering, 27 November 2011 through 29 November 2011, Tokushima ; 2011 , Pages 165-168 ; 9781612847283 (ISBN) ; Sameti, H ; Bokaei, M. H ; Chinese Association for Artificial Intelligence; IEEE Signal Processing Society ; Sharif University of Technology
2011
Abstract
The key element of a spoken dialogue system is Spoken Language Understanding (SLU) part. HVS and EHVS are two most popular statistical methods employed to implement the SLU part which need lightly annotated data. Since annotation is a time consuming, we present a novel semi-supervised learning for EHVS to reduce the human labeling effort using two different statistical classifiers, SVM and KNN. Experiments are done on a Persian corpus, the University Information Kiosk corpus. The experimental results show improvements in performance of semi-supervised EHVS, trained by both labeled and unlabeled data, compared to EHVS trained by just initially labeled data. The performance of EHVS improves...
Efficient iterative Semi-Supervised Classification on manifold
, Article Proceedings - IEEE International Conference on Data Mining, ICDM ; 2011 , Pages 228-235 ; 15504786 (ISSN); 9780769544090 (ISBN) ; Rabiee, H. R ; Shaban, A ; Soltani Farani, A ; National Science Foundation (NSF) - Where Discoveries Begin; University of Technology Sydney; Google; Alberta Ingenuity Centre for Machine Learning; IBM Research ; Sharif University of Technology
Abstract
Semi-Supervised Learning (SSL) has become a topic of recent research that effectively addresses the problem of limited labeled data. Many SSL methods have been developed based on the manifold assumption, among them, the Local and Global Consistency (LGC) is a popular method. The problem with most of these algorithms, and in particular with LGC, is the fact that their naive implementations do not scale well to the size of data. Time and memory limitations are the major problems faced in large-scale problems. In this paper, we provide theoretical bounds on gradient descent, and to overcome the aforementioned problems, a new approximate Newton's method is proposed. Moreover, convergence...