Forecasting market volatility, especially high-volatility incidents, is a critical issue in financial market research and practice. Business news as an important source of market information is often exploited by artificial intelligence–based volatility forecasting models. Computationally, deep learning architectures, such as recurrent neural networks, on extremely long input sequences remain infeasible because of time complexity and memory limitations. Meanwhile, understanding the inner workings of deep neural networks is challenging because of the largely black box nature of large neural networks. In this work, we address the first challenge by proposing a long- and short-term memory retrieval (LASER) architecture with flexible memory and horizon configurations to forecast market volatility. Then, we tackle the interpretability issue by devising a BEAM algorithm that leverages a large pretrained language model (GPT-2). It generates human-readable narratives verbalizing the evidence leading to the model prediction. Experiments on a Wall Street Journal news data set demonstrate the superior performance of our proposed LASER-BEAM pipeline in predicting high-volatility market scenarios and generating high-quality narratives compared with existing methods in the literature. History: Accepted by Ram Ramesh, Area Editor for Data Science & Machine Learning. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2022.0055 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2022.0055 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .

Existing studies in conversational AI mostly treat task-oriented dialog (TOD) and question answering (QA) as separate tasks. Towards the goal of constructing a conversational agent that can complete user tasks and support information seeking, it is important to develop a system that can handle both TOD and QA with access to various external knowledge sources. In this work, we propose a new task, Open-Book TOD (OB-TOD), which combines TOD with QA and expands the external knowledge sources to include both explicit sources (e.g., the web) and implicit sources (e.g., pre-trained language models). We create a new dataset OB-MultiWOZ, where we enrich TOD sessions with QA-like information-seeking experience grounded on external knowledge. We propose a unified model OPERA ( Op en-book E nd-to-end Task-o r iented Di a log) which can appropriately access explicit and implicit external knowledge to tackle the OB-TOD task. Experimental results show that OPERA outperforms closed-book baselines, highlighting the value of both types of knowledge. 1

The construction of dialog systems for various types of conversations, such as task-oriented dialog (TOD) and open-domain dialog (ODD), has been an active area of research.In order to more closely mimic human-like conversations that often involve the fusion of different dialog modes, it is important to develop systems that can effectively handle both TOD and ODD and access different knowledge sources.In this work, we present a new automatic framework to enrich TODs with synthesized ODDs.We also introduce the PivotBot model, which is capable of handling both TOD and ODD modes and can access different knowledge sources to generate informative responses.Evaluation results indicate the superior ability of the proposed model to switch smoothly between TOD and ODD tasks.

Topic models extract commonly occurring latent topics from textual data. Statistical models such as Latent Dirichlet Allocation do not produce dense topic embeddings readily integratable into neural architectures, whereas earlier neural topic models are yet to fully take advantage of the discrete nature of the topic space. To bridge this gap, we propose a novel neural topic model, Discrete-Variational-Inference-based Topic Model (DVITM), which learns dense topic embeddings homomorphic to word embeddings via discrete variational inference. The model also views words as mixtures of topics and digests embedded input text. Quantitative and qualitative evaluations empirically demonstrate the superior performance of DVITM compared to important baseline models. In the end, case studies on text generation from a discrete space and aspect-aware item recommendation are presented to further illustrate the power of our model in downstream tasks.

Detecting product adoption intentions on social media could yield significant value in a wide range of applications, such as personalized recommendations and targeted marketing. In the literature, no study has explored the detection of product adoption intentions on social media, and only a few relevant studies have focused on purchase intention detection for products in one or several categories. Focusing on a product category rather than a specific product is too coarse-grained for precise advertising. Additionally, existing studies primarily focus on using one type of text representation in target social media posts, ignoring the major yet unexplored potential of fusing different text representations. In this paper, we first formulate the problem of product adoption intention mining and demonstrate the necessity of studying this problem and its practical value. To detect a product adoption intention for an individual product, we propose a novel and general multiview deep learning model that simultaneously taps into the capability of multiview learning in leveraging different representations and deep learning in learning latent data representations using a flexible nonlinear transformation. Specifically, the proposed model leverages three different text representations from a multiview perspective and takes advantage of local and long-term word relations by integrating convolutional neural network (CNN) and long short-term memory (LSTM) modules. Extensive experiments on three Twitter datasets demonstrate the effectiveness of the proposed multiview deep learning model compared with the existing benchmark methods. This study also significantly contributes research insights to the literature about intention mining and provides business value to relevant stakeholders such as product providers.

Baolin Peng, Chunyuan Li, Zhu Zhang, Chenguang Zhu, Jinchao Li, Jianfeng Gao. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

With the purpose of learning and utilizing explicit and dense topic embeddings, we propose three variations of novel vector-quantization-based topic models (VQ-TMs): (1) Hard VQ-TM, (2) Soft VQ-TM, and (3) Multi-View Soft VQ-TM. The model family capitalize on vector quantization techniques, embedded input documents, and viewing words as mixtures of topics. Guided by a comprehensive set of evaluation metrics, we conduct systematic quantitative and qualitative empirical studies, and demonstrate the superior performance of VQ-TMs compared to important baseline models. Through a unique case study on code generation from natural language descriptions, we further illustrate the power of VQ-TMs in downstream tasks.

Label imbalance and data missing are two major challenges in the problem of credit card fraud detection. However, existing matrix completion algorithms are generally difficult and cannot be easily applied to real-world credit card fraud detection since the scale of the normally used dataset is oversized. In this paper, we develop a spectral regularization algorithm to complete the large-scale sparse matrices, and further utilize an over-sampling algorithm to tackle the problem of the imbalance between positive and negative samples. Experimental results on a real-world dataset demonstrate that our model can outperform the state-of-the-art baseline methods. The proposed method could also be extended to other large-scale scenarios where data is missing or labels are imbalanced.

In the context of deep learning models for semantic matching problems, we propose a novel Multi-View Progressive Attention (MV-PA) mechanism general enough to operate on various linguistic structures of text. More importantly, we study the interaction effect between explicit linguistic structures (e.g., linear, constituency, and dependency) and implicit structures elicited by attention mechanisms. Empirical results on multiple datasets demonstrate salient patterns of substitutability between the two families of structures (explicit and implicit). Our findings not only provide intellectual foundations for the popular use of “linear LSTM + attention” architectures in NLP/QA research, but also have implications in other modalities and domains.

How people divide their attention across their friends can help to understand key issues in the realm of geopolitics. Such attention exploration allows us to compare people who focus a large portion of their attention on a small set of close friends with those disperse their attention more widely. Using 2.5 million twitter data written by 130 thousands users, we find the balance of attention is a relatively stable property of people across different modalities of interaction. It displays subtle variation across people with different characteristics and different modalities of interaction. Specifically, people's attention is more focused in mention interactions, while those active in socialization tend to allocate higher portion of total attention to their close friends. Besides external interactions, people's inner interests also affect their attention allocation. People spreading multiple memes tend to be focused, and those with more even distribution of memes are focused on their intimate friends. Finally, people's relationships also plays an important role in their attention allocation. People are more likely to focus their attention on those most like them, and this similarity sequentially enhances the intimate relationship between them.

Discovering temporal patterns and changes in tobacco use has important practical implications in tobacco control. This paper presents one of the first comprehensive international studies of seasonal smoking patterns based on online searches performed. Using periodogram and cross-correlation, we find that smoking-related search behavior shows strong seasonality effect across countries. In addition, there are significant pairwise associations between such seasonality in different countries.