scholarly journals Profiling Users for Question Answering Communities via Flow-Based Constrained Co-Embedding Model

2022 ◽  
Vol 40 (2) ◽  
pp. 1-38
Shangsong Liang ◽  
Yupeng Luo ◽  
Zaiqiao Meng

In this article, we study the task of user profiling in question answering communities (QACs). Previous user profiling algorithms suffer from a number of defects: they regard users and words as atomic units, leading to the mismatch between them; they are designed for other applications but not for QACs; and some semantic profiling algorithms do not co-embed users and words, leading to making the affinity measurement between them difficult. To improve the profiling performance, we propose a neural Flow-based Constrained Co-embedding Model, abbreviated as FCCM. FCCM jointly co-embeds the vector representations of both users and words in QACs such that the affinities between them can be semantically measured. Specifically, FCCM extends the standard variational auto-encoder model to enforce the inferred embeddings of users and words subject to the voting constraint, i.e., given a question and the users who answer this question in the community, representations of the users whose answers receive more votes are closer to the representations of the words associated with these answers, compared with representations of whose receiving fewer votes. In addition, FCCM integrates normalizing flow into the variational auto-encoder framework to avoid the assumption that the distributions of the embeddings are Gaussian, making the inferred embeddings fit the real distributions of the data better. Experimental results on a Chinese Zhihu question answering dataset demonstrate the effectiveness of our proposed FCCM model for the task of user profiling in QACs.

2021 ◽  
Vol 11 (2) ◽  
pp. 721
Hyung Yong Kim ◽  
Ji Won Yoon ◽  
Sung Jun Cheon ◽  
Woo Hyun Kang ◽  
Nam Soo Kim

Recently, generative adversarial networks (GANs) have been successfully applied to speech enhancement. However, there still remain two issues that need to be addressed: (1) GAN-based training is typically unstable due to its non-convex property, and (2) most of the conventional methods do not fully take advantage of the speech characteristics, which could result in a sub-optimal solution. In order to deal with these problems, we propose a progressive generator that can handle the speech in a multi-resolution fashion. Additionally, we propose a multi-scale discriminator that discriminates the real and generated speech at various sampling rates to stabilize GAN training. The proposed structure was compared with the conventional GAN-based speech enhancement algorithms using the VoiceBank-DEMAND dataset. Experimental results showed that the proposed approach can make the training faster and more stable, which improves the performance on various metrics for speech enhancement.

Robotics ◽  
2021 ◽  
Vol 10 (2) ◽  
pp. 68
Lei Shi ◽  
Cosmin Copot ◽  
Steve Vanlanduit

In gaze-based Human-Robot Interaction (HRI), it is important to determine human visual intention for interacting with robots. One typical HRI interaction scenario is that a human selects an object by gaze and a robotic manipulator will pick up the object. In this work, we propose an approach, GazeEMD, that can be used to detect whether a human is looking at an object for HRI application. We use Earth Mover’s Distance (EMD) to measure the similarity between the hypothetical gazes at objects and the actual gazes. Then, the similarity score is used to determine if the human visual intention is on the object. We compare our approach with a fixation-based method and HitScan with a run length in the scenario of selecting daily objects by gaze. Our experimental results indicate that the GazeEMD approach has higher accuracy and is more robust to noises than the other approaches. Hence, the users can lessen cognitive load by using our approach in the real-world HRI scenario.

2021 ◽  
pp. 1-12
Melesio Crespo-Sanchez ◽  
Ivan Lopez-Arevalo ◽  
Edwin Aldana-Bobadilla ◽  
Alejandro Molina-Villegas

In the last few years, text analysis has grown as a keystone in several domains for solving many real-world problems, such as machine translation, spam detection, and question answering, to mention a few. Many of these tasks can be approached by means of machine learning algorithms. Most of these algorithms take as input a transformation of the text in the form of feature vectors containing an abstraction of the content. Most of recent vector representations focus on the semantic component of text, however, we consider that also taking into account the lexical and syntactic components the abstraction of content could be beneficial for learning tasks. In this work, we propose a content spectral-based text representation applicable to machine learning algorithms for text analysis. This representation integrates the spectra from the lexical, syntactic, and semantic components of text producing an abstract image, which can also be treated by both, text and image learning algorithms. These components came from feature vectors of text. For demonstrating the goodness of our proposal, this was tested on text classification and complexity reading score prediction tasks obtaining promising results.

2021 ◽  
Vol 9 ◽  
pp. 929-944
Omar Khattab ◽  
Christopher Potts ◽  
Matei Zaharia

Abstract Systems for Open-Domain Question Answering (OpenQA) generally depend on a retriever for finding candidate passages in a large corpus and a reader for extracting answers from those passages. In much recent work, the retriever is a learned component that uses coarse-grained vector representations of questions and passages. We argue that this modeling choice is insufficiently expressive for dealing with the complexity of natural language questions. To address this, we define ColBERT-QA, which adapts the scalable neural retrieval model ColBERT to OpenQA. ColBERT creates fine-grained interactions between questions and passages. We propose an efficient weak supervision strategy that iteratively uses ColBERT to create its own training data. This greatly improves OpenQA retrieval on Natural Questions, SQuAD, and TriviaQA, and the resulting system attains state-of-the-art extractive OpenQA performance on all three datasets.

2010 ◽  
Vol 121-122 ◽  
pp. 43-47 ◽  
Li Ying Wang ◽  
Wei Guo Zhao

Relevance Vector Machine (RVM) is a novel kernel method based on sparse Bayesian, which has many advantages such as its kernel functions without the restriction of Mercer’s conditions, and the relevance vectors are automatically determined and have fewer parameters. In this paper, the RVM model is applied to forecasting groundwater level. The experimental results show the final RVM model achieved is sparser, the prediction precision is higher and the prediction values are in better agreement with the real values. It can be concluded that this technique can be seen as a very promising option to solve nonlinear problems such as forecasting groundwater level.

Juan Zhang ◽  
Wenbin Guo

This article propose s a network that is mainly used to deal with a single image polluted by raindrops in rainy weather to get a clean image without raindrops. In the existing solutions, most of the methods rely on paired images, that is, the rain image and the real image without rain in the same scene. However, in many cases, the paired images are difficult to obtain, which makes it impossible to apply the raindrop removal network in many scenarios. Therefore this article proposes a semi-supervised rain-removing network apply to unpaired images. The model contains two parts: a supervised network and an unsupervised network. After the model is trained, the unsupervised network does not require paired images and it can get a clean image without raindrops. In particular, our network can perform training on paired and unpaired samples. The experimental results show that the best results are achieved not only on the supervised rain-removing network, but also on the unsupervised rain-removing network.

Rafal Rzepka ◽  
Kenji Araki

This chapter introduces an approach and methods for creating a system that refers to human experiences and thoughts about these experiences in order to ethically evaluate other parties', and in a long run, its own actions. It is shown how applying text mining techniques can enrich machine's knowledge about the real world and how this knowledge could be helpful in the difficult realm of moral relativity. Possibilities of simulating empathy and applying proposed methods to various approaches are introduced together with discussion on the possibility of applying growing knowledge base to artificial agents for particular purposes, from simple housework robots to moral advisors, which could refer to millions of different experiences had by people in various cultures. The experimental results show efficiency improvements when compared to previous research and also discuss the problems with fair evaluation of moral and immoral acts.

Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-15 ◽  
Hanwen Liu ◽  
Huaizhen Kou ◽  
Chao Yan ◽  
Lianyong Qi

Nowadays, scholar recommender systems often recommend academic papers based on users’ personalized retrieval demands. Typically, a recommender system analyzes the keywords typed by a user and then returns his or her preferred papers, in an efficient and economic manner. In practice, one paper often contains partial keywords that a user is interested in. Therefore, the recommender system needs to return the user a set of papers that collectively covers all the queried keywords. However, existing recommender systems only use the exact keyword matching technique for recommendation decisions, while neglecting the correlation relationships among different papers. As a consequence, it may output a set of papers from multiple disciplines that are different from the user’s real research field. In view of this shortcoming, we propose a keyword-driven and popularity-aware paper recommendation approach based on an undirected paper citation graph, named PRkeyword+pop. At last, we conduct large-scale experiments on the real-life Hep-Th dataset to further demonstrate the usefulness and feasibility of PRkeyword+pop. Experimental results prove the advantages of PRkeyword+pop in searching for a set of satisfactory papers compared with other competitive approaches.

2020 ◽  
Vol 34 (05) ◽  
pp. 9330-9337
Dong Xu ◽  
Wu-Jun Li

Answer selection is an important subtask of question answering (QA), in which deep models usually achieve better performance than non-deep models. Most deep models adopt question-answer interaction mechanisms, such as attention, to get vector representations for answers. When these interaction based deep models are deployed for online prediction, the representations of all answers need to be recalculated for each question. This procedure is time-consuming for deep models with complex encoders like BERT which usually have better accuracy than simple encoders. One possible solution is to store the matrix representation (encoder output) of each answer in memory to avoid recalculation. But this will bring large memory cost. In this paper, we propose a novel method, called hashing based answer selection (HAS), to tackle this problem. HAS adopts a hashing strategy to learn a binary matrix representation for each answer, which can dramatically reduce the memory cost for storing the matrix representations of answers. Hence, HAS can adopt complex encoders like BERT in the model, but the online prediction of HAS is still fast with a low memory cost. Experimental results on three popular answer selection datasets show that HAS can outperform existing models to achieve state-of-the-art performance.

2013 ◽  
Vol 816-817 ◽  
pp. 488-492
Li Xin Li ◽  
Wei Zhou ◽  
Qi Qiang Sun ◽  
Jiao Dai ◽  
Ji Zhong Han ◽  

In order to make the real time database more suitable for the computing features, this article points to the distributed and parallel real time database design and architecture. First, a mapping table from table file to machine nodes is established, and then can use meta-data management system to store and manage the mapping table to meet the characteristics of high concurrent access. The whole network computation can access the unified interface provided by the real-time database, retrive data from each node, then collect the data. Experimental results show that this study and the systems designed can meet the computing requirements of a unified whole network.

Sign in / Sign up

Export Citation Format

Share Document