Sarcasm is the use of words usually used to either mock or annoy someone, or for humorous purposes. Sarcasm is largely used in social networks and microblogging websites, where people mock or censure in a way that makes it difficult even for humans to tell if what is said is what is meant. Failure to identify sarcastic utterances in Natural Language Processing applications such as sentiment analysis and opinion mining will confuse classification algorithms and generate false results. Several studies on sarcasm detection have utilized different learning algorithms. However, most of these learning models have always focused on the contents of expression only, leaving the contextual information in isolation. As a result, they failed to capture the contextual information in the sarcastic expression. Moreover, some datasets used in several studies have an unbalanced dataset which impacting the model result. In this paper, we propose a contextual model for sarcasm identification in twitter using RoBERTa, and augmenting the dataset by applying Global Vector representation (GloVe) for the construction of word embedding and context learning to generate more data and balancing the dataset. The effectiveness of this technique is tested with various datasets and data augmentation settings. In particular, we achieve performance gain by 3.2% in the iSarcasm dataset when using data augmentation to increase 20% of data labeled as sarcastic, resulting F-score of 40.4% compared to 37.2% without data augmentation.

We study the problem of query attribute value extraction, which aims to identify named entities from user queries as diverse surface form attribute values and afterward transform them into formally canonical forms. Such a problem consists of two phases: {named entity recognition (NER)} and {attribute value normalization (AVN)}. However, existing works only focus on the NER phase but neglect equally important AVN. To bridge this gap, this paper proposes a unified query attribute value extraction system in e-commerce search named QUEACO, which involves both two phases. Moreover, by leveraging large-scale weakly-labeled behavior data, we further improve the extraction performance with less supervision cost. Specifically, for the NER phase, QUEACO adopts a novel teacher-student network, where a teacher network that is trained on the strongly-labeled data generates pseudo-labels to refine the weakly-labeled data for training a student network. Meanwhile, the teacher network can be dynamically adapted by the feedback of the student's performance on strongly-labeled data to maximally denoise the noisy supervisions from the weak labels. For the AVN phase, we also leverage the weakly-labeled query-to-attribute behavior data to normalize surface form attribute values from queries into canonical forms from products. Extensive experiments on a real-world large-scale E-commerce dataset demonstrate the effectiveness of QUEACO.

The GoEmotions dataset contains 58k carefully curated Reddit comments labeled for 27 emotion categories or Neutral. The raw data is included as well as the smaller, simplified version of the dataset with predefined train/val/test splits. After going through a few examples in this dataset on their visualizer, I realized that this is an extremely crucial dataset because it's rare to find sentiment classifier datasets that go beyond 5–6 emotions. But here, we have 27 emotions being assigned, with rare and close enough emotions like disappointment, disapproval, grief, remorse, sadness, etc. Detecting such close enough emotions is often difficult in typical datasets. This made it clear to me that this is an excellent dataset that can be scaled for usage in many applications that involve text analysis.

But the debates between Facebook and the White House throughout the spring and into summer also gave rise to a broader and still unresolved disagreement over what constitutes misinformation, according to the person familiar with Facebook's thinking. Facebook strongly believes people should have the right to broadly express themselves without censorship on social platforms, and had reviewed research that shows that friends and family can often be more effective at countering misinformation than official sources that people distrust. The Facebook executives thought the Biden camp was going too far, by identifying specific pieces of content as problematic and asking it to potentially suppress valuable conversations where people express fears and skepticism.

The multi-format information extraction task in the 2021 Language and Intelligence Challenge is designed to comprehensively evaluate information extraction from different dimensions. It consists of an multiple slots relation extraction subtask and two event extraction subtasks that extract events from both sentence-level and document-level. Here we describe our system for this multi-format information extraction competition task. Specifically, for the relation extraction subtask, we convert it to a traditional triple extraction task and design a voting based method that makes full use of existing models. For the sentence-level event extraction subtask, we convert it to a NER task and use a pointer labeling based method for extraction. Furthermore, considering the annotated trigger information may be helpful for event extraction, we design an auxiliary trigger recognition model and use the multi-task learning mechanism to integrate the trigger features into the event extraction model. For the document-level event extraction subtask, we design an Encoder-Decoder based method and propose a Transformer-alike decoder. Finally,our system ranks No.4 on the test set leader-board of this multi-format information extraction task, and its F1 scores for the subtasks of relation extraction, event extractions of sentence-level and document-level are 79.887%, 85.179%, and 70.828% respectively. The codes of our model are available at {https://github.com/neukg/MultiIE}.

I read Modern Romance by Aziz Ansari in 2016 and beyond a shadow of a doubt, it is one of the most influential books I've ever read. At the time, I was a snot-nosed college student who was still dating someone from high school. The numbers and figures given by the book about online dating success struck me as being callous. Millennials and their predecessors were blessed and cursed with the advent of the internet. The proliferation of partner-choice desensitizes us and gives us unrealistic expectations when it came to searching for our "soulmate." Instead of feeling dissuaded, I felt inspired.

Aspect Sentiment Triplet Extraction (ASTE) is the task of extracting triplets of aspect terms, their associated sentiments, and the opinion terms that provide evidence for the expressed sentiments. Previous approaches to ASTE usually simultaneously extract all three components or first identify the aspect and opinion terms, then pair them up to predict their sentiment polarities. In this work, we present a novel paradigm, ASTE-RL, by regarding the aspect and opinion terms as arguments of the expressed sentiment in a hierarchical reinforcement learning (RL) framework. We first focus on sentiments expressed in a sentence, then identify the target aspect and opinion terms for that sentiment. This takes into account the mutual interactions among the triplet's components while improving exploration and sample efficiency. Furthermore, this hierarchical RLsetup enables us to deal with multiple and overlapping triplets. In our experiments, we evaluate our model on existing datasets from laptop and restaurant domains and show that it achieves state-of-the-art performance. The implementation of this work is publicly available at https://github.com/declare-lab/ASTE-RL.

We present VESPA, an intentionally simple yet novel zero-shot system for layout, locale, and domain agnostic document extraction. In spite of the availability of large corpora of documents, the lack of labeled and validated datasets makes it a challenge to discriminatively train document extraction models for enterprises. We show that this problem can be addressed by simply transferring the information extraction (IE) task to a natural language Question-Answering (QA) task without engineering task-specific architectures. We demonstrate the effectiveness of our system by evaluating on a closed corpus of real-world retail and tax invoices with multiple complex layouts, domains, and geographies. The empirical evaluation shows that our system outperforms 4 prominent commercial invoice solutions that use discriminatively trained models with architectures specifically crafted for invoice extraction. We extracted 6 fields with zero upfront human annotation or training with an Avg. F1 of 87.50.

Facebook has today announced that it has added two new destinations for when you want to move your data from the social network. In a blog post, the company said that users will be able to move their images to Photobucket and event listings to Google Calendar. Product Manager Hadi Michel said that the tool has been "completely rebuilt" to be "simpler and more intuitive," giving people more clarity on what they can share to which platforms. In addition, users can now launch multiple transfers, with better fine-grain control on what they're choosing to export in any one transfer. This is yet another feature piled on to the Data Transfer Project, an open-source project developed by Google, Facebook and Microsoft.

India witnessed its first-ever nationwide lockdown from 24th March 2020 to 31st May 2020 to fight the spread of the novel coronavirus by limiting the movement of its residents. The study of this article aims at identifying the different emotions of customers in ordering pizza in India from one of the most popular pizza delivery chains called Dominos. From 128 stores in 2006 Domino's India, the poster brand of Jubilant Foodworks reported more than 1300 stores in 2020 across the country. The study analyzed Twitter data during three different periods. The first period from 1st January 2020 to 24th March 2020 was considered as a pre-lockdown, the second period from 25th March 2020 to 31st May 2020 was considered as the lockdown period and finally, the third period from 1st January 2021 to 28th February 2021 was considered as post-lockdown. Twitter data for these three different periods were extracted using Sprinklr (licensed by IIM Ahmedabad), which is specialized in providing real-time user conversations from modern social handles.