Power-efficient CNN Domain Specific Accelerator (CNN-DSA) chips are currently available for wide use in mobile devices. These chips are mainly used in computer vision applications. However, the recent work of Super Characters method for text classification and sentiment analysis tasks using two-dimensional CNN models has also achieved state-of-the-art results through the method of transfer learning from vision to text. In this paper, we implemented the text classification and sentiment analysis applications on mobile devices using CNN-DSA chips. Compact network representations using one-bit and three-bits precision for coefficients and five-bits for activations are used in the CNN-DSA chip with power consumption less than 300mW. For edge devices under memory and compute constraints, the network is further compressed by approximating the external Fully Connected (FC) layers within the CNN-DSA chip. At the workshop, we have two system demonstrations for NLP tasks. The first demo classifies the input English Wikipedia sentence into one of the 14 ontologies. The second demo classifies the Chinese online-shopping review into positive or negative.
Domain Adaptation explores the idea of how to maximize performance on a target domain, distinct from source domain, upon which the classifier was trained. This idea has been explored for the task of sentiment analysis extensively. The training of reviews pertaining to one domain and evaluation on another domain is widely studied for modeling a domain independent algorithm. This further helps in understanding correlation between domains. In this paper, we show that Gated Convolutional Neural Networks (GCN) perform effectively at learning sentiment analysis in a manner where domain dependant knowledge is filtered out using its gates. We perform our experiments on multiple gate architectures: Gated Tanh ReLU Unit (GTRU), Gated Tanh Unit (GTU) and Gated Linear Unit (GLU). Extensive experimentation on two standard datasets relevant to the task, reveal that training with Gated Convolutional Neural Networks give significantly better performance on target domains than regular convolution and recurrent based architectures. While complex architectures like attention, filter domain specific knowledge as well, their complexity order is remarkably high as compared to gated architectures. GCNs rely on convolution hence gaining an upper hand through parallelization.
Text classification has long been a staple in natural language processing with applications spanning across sentiment analysis, online content tagging, recommender systems and spam detection. However, text classification, by nature, suffers from a variety of issues stemming from dataset imbalance, text ambiguity, subjectivity and the lack of linguistic context in the data. In this paper, we explore the use of text ranking, commonly used in information retrieval, to carry out challenging classification-based tasks. We propose a novel end-to-end ranking approach consisting of a Transformer network responsible for producing representations for a pair of text sequences, which are in turn passed into a context aggregating network outputting ranking scores used to determine an ordering to the sequences based on some notion of relevance. We perform numerous experiments on publicly-available datasets and investigate the possibility of applying our ranking approach to certain problems often addressed using classification. In an experiment on a heavily-skewed sentiment analysis dataset, converting ranking results to classification labels yields an approximately 22% improvement over state-of-the-art text classification, demonstrating the efficacy of text ranking over text classification in certain scenarios.
Multimodal language analysis often considers relationships between features based on text and those based on acoustical and visual properties. Text features typically outperform non-text features in sentiment analysis or emotion recognition tasks in part because the text features are derived from advanced language models or word embeddings trained on massive data sources while audio and video features are human-engineered and comparatively underdeveloped. Given that the text, audio, and video are describing the same utterance in different ways, we hypothesize that the multimodal sentiment analysis and emotion recognition can be improved by learning (hidden) correlations between features extracted from the outer product of text and audio (we call this text-based audio) and analogous text-based video. This paper proposes a novel model, the Interaction Canonical Correlation Network (ICCN), to learn such multimodal embeddings. ICCN learns correlations between all three modes via deep canonical correlation analysis (DCCA) and the proposed embeddings are then tested on several benchmark datasets and against other state-of-the-art multimodal embedding algorithms. Empirical results and ablation studies confirm the effectiveness of ICCN in capturing useful information from all three views.
Sentence embedding is an important research topic in natural language processing. It is essential to generate a good embedding vector that fully reflects the semantic meaning of a sentence in order to achieve an enhanced performance for various natural language processing tasks, such as machine translation and document classification. Thus far, various sentence embedding models have been proposed, and their feasibility has been demonstrated through good performances on tasks following embedding, such as sentiment analysis and sentence classification. However, because the performances of sentence classification and sentiment analysis can be enhanced by using a simple sentence representation method, it is not sufficient to claim that these models fully reflect the meanings of sentences based on good performances for such tasks. In this paper, inspired by human language recognition, we propose the following concept of semantic coherence, which should be satisfied for a good sentence embedding method: similar sentences should be located close to each other in the embedding space. Then, we propose the Paraphrase-Thought (P-thought) model to pursue semantic coherence as much as possible. Experimental results on two paraphrase identification datasets (MS COCO and STS benchmark) show that the P-thought models outperform the benchmarked sentence embedding methods.
In this article we propose a new supervised ensemble learning method called Data Shared Adaptive Bootstrap Aggregated (AdaBag) Lasso for capturing low dimensional useful features for word based sentiment analysis and mining problems. The literature on ensemble methods is very rich in both statistics and machine learning. The algorithm is a substantial upgrade of the Data Shared Lasso uplift algorithm. The most significant conceptual addition to the existing literature lies in the final selection of bag of predictors through a special bootstrap aggregation scheme. We apply the algorithm to one simulated data and perform dimension reduction in grouped IMDb data (drama, comedy and horror) to extract reduced set of word features for predicting sentiment ratings of movie reviews demonstrating different aspects. We also compare the performance of the present method with the classical Principal Components with associated Linear Discrimination (PCA-LD) as baseline. There are few limitations in the algorithm. Firstly, the algorithm workflow does not incorporate online sequential data acquisition and it does not use sentence based models which are common in ANN algorithms . Our results produce slightly higher error rate compare to the reported state-of-the-art as a consequence.
One of the most important incidents in the world in 2020 is the outbreak of the Coronavirus. Users on social networks publish a large number of comments about this event. These comments contain important hidden information of public opinion regarding this pandemic. In this research, a large number of Coronavirus-related tweets are considered and analyzed using natural language processing and information retrieval science. Initially, the location of the tweets is determined using a dictionary prepared through the Geo-Names geographic database, which contains detailed and complete information of places such as city names, streets, and postal codes. Then, using a large dictionary prepared from the terms of economics, related tweets are extracted and sentiments corresponded to tweets are analyzed with the help of the RoBERTa language-based model, which has high accuracy and good performance. Finally, the frequency chart of tweets related to the economy and their sentiment scores (positive and negative tweets) is plotted over time for the entire world and the top 10 economies. From the analysis of the charts, we learn that the reason for publishing economic tweets is not only the increase in the number of people infected with the Coronavirus but also imposed restrictions and lockdowns in countries. The consequences of these restrictions include the loss of millions of jobs and the economic downturn.
Although self-attention based models such as Transformers have achieved remarkable successes on natural language processing (NLP) tasks, recent studies reveal that they have limitations on modeling sequential transformations (Hahn, 2020), which may prompt re-examinations of recurrent neural networks (RNNs) that demonstrated impressive results on handling sequential data. Despite many prior attempts to interpret RNNs, their internal mechanisms have not been fully understood, and the question on how exactly they capture sequential features remains largely unclear. In this work, we present a study that shows there actually exist some explainable components that reside within the hidden states, which are reminiscent of the classical n-grams features. We evaluated such extracted explainable features from trained RNNs on downstream sentiment analysis tasks and found they could be used to model interesting linguistic phenomena such as negation and intensification. Furthermore, we examined the efficacy of using such n-gram components alone as encoders on tasks such as sentiment analysis and language modeling, revealing they could be playing important roles in contributing to the overall performance of RNNs. We hope our findings could add interpretability to RNN architectures, and also provide inspirations for proposing new architectures for sequential data.