Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

"Sentiment": models, code, and papers

DP-LSTM: Differential Privacy-inspired LSTM for Stock Prediction Using Financial News

Dec 20, 2019
Xinyi Li, Yinchuan Li, Hongyang Yang, Liuqing Yang, Xiao-Yang Liu

Stock price prediction is important for value investments in the stock market. In particular, short-term prediction that exploits financial news articles is promising in recent years. In this paper, we propose a novel deep neural network DP-LSTM for stock price prediction, which incorporates the news articles as hidden information and integrates difference news sources through the differential privacy mechanism. First, based on the autoregressive moving average model (ARMA), a sentiment-ARMA is formulated by taking into consideration the information of financial news articles in the model. Then, an LSTM-based deep neural network is designed, which consists of three components: LSTM, VADER model and differential privacy (DP) mechanism. The proposed DP-LSTM scheme can reduce prediction errors and increase the robustness. Extensive experiments on S&P 500 stocks show that (i) the proposed DP-LSTM achieves 0.32% improvement in mean MPA of prediction result, and (ii) for the prediction of the market index S&P 500, we achieve up to 65.79% improvement in MSE.

* arXiv admin note: text overlap with arXiv:1908.01112 

  Access Paper or Ask Questions

RNN Architecture Learning with Sparse Regularization

Sep 06, 2019
Jesse Dodge, Roy Schwartz, Hao Peng, Noah A. Smith

Neural models for NLP typically use large numbers of parameters to reach state-of-the-art performance, which can lead to excessive memory usage and increased runtime. We present a structure learning method for learning sparse, parameter-efficient NLP models. Our method applies group lasso to rational RNNs (Peng et al., 2018), a family of models that is closely connected to weighted finite-state automata (WFSAs). We take advantage of rational RNNs' natural grouping of the weights, so the group lasso penalty directly removes WFSA states, substantially reducing the number of parameters in the model. Our experiments on a number of sentiment analysis datasets, using both GloVe and BERT embeddings, show that our approach learns neural structures which have fewer parameters without sacrificing performance relative to parameter-rich baselines. Our method also highlights the interpretable properties of rational RNNs. We show that sparsifying such models makes them easier to visualize, and we present models that rely exclusively on as few as three WFSAs after pruning more than 90% of the weights. We publicly release our code.

  Access Paper or Ask Questions

Certified Robustness to Adversarial Word Substitutions

Sep 03, 2019
Robin Jia, Aditi Raghunathan, Kerem Göksel, Percy Liang

State-of-the-art NLP models can often be fooled by adversaries that apply seemingly innocuous label-preserving transformations (e.g., paraphrasing) to input text. The number of possible transformations scales exponentially with text length, so data augmentation cannot cover all transformations of an input. This paper considers one exponentially large family of label-preserving transformations, in which every word in the input can be replaced with a similar word. We train the first models that are provably robust to all word substitutions in this family. Our training procedure uses Interval Bound Propagation (IBP) to minimize an upper bound on the worst-case loss that any combination of word substitutions can induce. To evaluate models' robustness to these transformations, we measure accuracy on adversarially chosen word substitutions applied to test examples. Our IBP-trained models attain $75\%$ adversarial accuracy on both sentiment analysis on IMDB and natural language inference on SNLI. In comparison, on IMDB, models trained normally and ones trained with data augmentation achieve adversarial accuracy of only $8\%$ and $35\%$, respectively.

* EMNLP 2019 

  Access Paper or Ask Questions

StructBERT: Incorporating Language Structures into Pre-training for Deep Language Understanding

Aug 16, 2019
Wei Wang, Bin Bi, Ming Yan, Chen Wu, Zuyi Bao, Liwei Peng, Luo Si

Recently, the pre-trained language model, BERT, has attracted a lot of attention in natural language understanding (NLU), and achieved state-of-the-art accuracy in various NLU tasks, such as sentiment classification, natural language inference, semantic textual similarity and question answering. Inspired by the linearization exploration work of Elman, we extend BERT to a new model, StructBERT, by incorporating language structures into pre-training. Specifically, we pre-train StructBERT with two auxiliary tasks to make the most of the sequential order of words and sentences, which leverage language structures at the word and sentence levels, respectively. As a result, the new model is adapted to different levels of language understanding required by downstream tasks. The StructBERT with structural pre-training gives surprisingly good empirical results on a variety of downstream tasks, including pushing the state-of-the-art on the GLUE benchmark to 84.5 (with Top 1 achievement on the Leaderboard at the time of paper submission), the F1 score on SQuAD v1.1 question answering to 93.0, the accuracy on SNLI to 91.7.

* 10 Pages 

  Access Paper or Ask Questions

M-BERT: Injecting Multimodal Information in the BERT Structure

Aug 15, 2019
Wasifur Rahman, Md Kamrul Hasan, Amir Zadeh, Louis-Philippe Morency, Mohammed Ehsan Hoque

Multimodal language analysis is an emerging research area in natural language processing that models language in a multimodal manner. It aims to understand language from the modalities of text, visual, and acoustic by modeling both intra-modal and cross-modal interactions. BERT (Bidirectional Encoder Representations from Transformers) provides strong contextual language representations after training on large-scale unlabeled corpora. Fine-tuning the vanilla BERT model has shown promising results in building state-of-the-art models for diverse NLP tasks like question answering and language inference. However, fine-tuning BERT in the presence of information from other modalities remains an open research problem. In this paper, we inject multimodal information within the input space of BERT network for modeling multimodal language. The proposed injection method allows BERT to reach a new state of the art of $84.38\%$ binary accuracy on CMU-MOSI dataset (multimodal sentiment analysis) with a gap of 5.98 percent to the previous state of the art and 1.02 percent to the text-only BERT.

  Access Paper or Ask Questions

Twitter Speaks: A Case of National Disaster Situational Awareness

Mar 07, 2019
Amir Karami, Vishal Shah, Reza Vaezi, Amit Bansal

In recent years, we have been faced with a series of natural disasters causing a tremendous amount of financial, environmental, and human losses. The unpredictable nature of natural disasters' behavior makes it hard to have a comprehensive situational awareness (SA) to support disaster management. Using opinion surveys is a traditional approach to analyze public concerns during natural disasters; however, this approach is limited, expensive, and time-consuming. Luckily the advent of social media has provided scholars with an alternative means of analyzing public concerns. Social media enable users (people) to freely communicate their opinions and disperse information regarding current events including natural disasters. This research emphasizes the value of social media analysis and proposes an analytical framework: Twitter Situational Awareness (TwiSA). This framework uses text mining methods including sentiment analysis and topic modeling to create a better SA for disaster preparedness, response, and recovery. TwiSA has also effectively deployed on a large number of tweets and tracks the negative concerns of people during the 2015 South Carolina flood.

* 17 pages, 3 figures, 5 tables 

  Access Paper or Ask Questions

Visualizing and Understanding Neural Models in NLP

Jan 08, 2016
Jiwei Li, Xinlei Chen, Eduard Hovy, Dan Jurafsky

While neural networks have been successfully applied to many NLP tasks the resulting vector-based models are very difficult to interpret. For example it's not clear how they achieve {\em compositionality}, building sentence meaning from the meanings of words and phrases. In this paper we describe four strategies for visualizing compositionality in neural models for NLP, inspired by similar work in computer vision. We first plot unit values to visualize compositionality of negation, intensification, and concessive clauses, allow us to see well-known markedness asymmetries in negation. We then introduce three simple and straightforward methods for visualizing a unit's {\em salience}, the amount it contributes to the final composed meaning: (1) gradient back-propagation, (2) the variance of a token from the average word node, (3) LSTM-style gates that measure information flow. We test our methods on sentiment using simple recurrent nets and LSTMs. Our general-purpose methods may have wide applications for understanding compositionality and other semantic properties of deep networks , and also shed light on why LSTMs outperform simple recurrent nets,

  Access Paper or Ask Questions

Polarity detection movie reviews in hindi language

Sep 13, 2014
Richa Sharma, Shweta Nigam, Rekha Jain

Nowadays peoples are actively involved in giving comments and reviews on social networking websites and other websites like shopping websites, news websites etc. large number of people everyday share their opinion on the web, results is a large number of user data is collected .users also find it trivial task to read all the reviews and then reached into the decision. It would be better if these reviews are classified into some category so that the user finds it easier to read. Opinion Mining or Sentiment Analysis is a natural language processing task that mines information from various text forms such as reviews, news, and blogs and classify them on the basis of their polarity as positive, negative or neutral. But, from the last few years, user content in Hindi language is also increasing at a rapid rate on the Web. So it is very important to perform opinion mining in Hindi language as well. In this paper a Hindi language opinion mining system is proposed. The system classifies the reviews as positive, negative and neutral for Hindi language. Negation is also handled in the proposed system. Experimental results using reviews of movies show the effectiveness of the system

  Access Paper or Ask Questions

Singer separation for karaoke content generation

Oct 13, 2021
Hsuan-Yu Chen, Xuanjun Chen, Jyh-Shing Roger Jang

Due to the rapid development of deep learning, we can now successfully separate singing voice from mono audio music. However, this separation can only extract human voices from other musical instruments, which is undesirable for karaoke content generation applications that only require the separation of lead singers. For this karaoke application, we need to separate the music containing male and female duets into two vocals, or extract a single lead vocal from the music containing vocal harmony. For this reason, we propose in this article to use a singer separation system, which generates karaoke content for one or two separated lead singers. In particular, we introduced three models for the singer separation task and designed an automatic model selection scheme to distinguish how many lead singers are in the song. We also collected a large enough data set, MIR-SingerSeparation, which has been publicly released to advance the frontier of this research. Our singer separation is most suitable for sentimental ballads and can be directly applied to karaoke content generation. As far as we know, this is the first singer-separation work for real-world karaoke applications.

* Submitted to ICASSP 2022 

  Access Paper or Ask Questions

Graph Capsule Aggregation for Unaligned Multimodal Sequences

Aug 17, 2021
Jianfeng Wu, Sijie Mai, Haifeng Hu

Humans express their opinions and emotions through multiple modalities which mainly consist of textual, acoustic and visual modalities. Prior works on multimodal sentiment analysis mostly apply Recurrent Neural Network (RNN) to model aligned multimodal sequences. However, it is unpractical to align multimodal sequences due to different sample rates for different modalities. Moreover, RNN is prone to the issues of gradient vanishing or exploding and it has limited capacity of learning long-range dependency which is the major obstacle to model unaligned multimodal sequences. In this paper, we introduce Graph Capsule Aggregation (GraphCAGE) to model unaligned multimodal sequences with graph-based neural model and Capsule Network. By converting sequence data into graph, the previously mentioned problems of RNN are avoided. In addition, the aggregation capability of Capsule Network and the graph-based structure enable our model to be interpretable and better solve the problem of long-range dependency. Experimental results suggest that GraphCAGE achieves state-of-the-art performance on two benchmark datasets with representations refined by Capsule Network and interpretation provided.

  Access Paper or Ask Questions