Indoor localization technology, a core component of applications such as smart homes and the Internet of Things (IoT), has attracted considerable attention in recent years. Fingerprint-based localization relies on the construction of dense, high-quality fingerprint databases but faces two major challenges: 1) the time-consuming and complex process of collecting received signal strength indicator (RSSI) fingerprint samples and 2) the significant fluctuations in fingerprint data caused by indoor environmental factors, which lead to inaccurate matching. These challenges affect the accuracy, stability, and scalability of localization systems. To address these issues, this article proposes a sparse fingerprint sample collection strategy and introduces a novel few-shot indoor localization model based on a simplified graph convolutional and adversarial Gaussian process regression (SGC-AGPR). The proposed approach reduces the sample collection burden by using sparse sampling, aggregates feature from neighboring nodes through a simplified graph convolutional model (SGC), synthesizes fingerprints for synthesized reference points (SRPs), and incorporates the spatial topological relationships of the reference points (RPs). Additionally, an optimized adversarial Gaussian process regression model (AGPR) is employed to provide initial values for SRP within the graph convolution process, enhancing the statistical correlation between fingerprint samples. Furthermore, this method introduces similarity-based adaptive weights to determine aggregation weights, mitigating issues related to inaccurate fingerprint matching. Experimental results demonstrate that the proposed localization model achieves an accuracy of approximately 0.84 m with limited samples, offering significant improvements in cost, accuracy, and stability compared to traditional methods. The proposed approach provides an innovative solution to the challenges of limited sample availability and environmental interference, contributing to the advancement of indoor localization technology.

Mobile technology is being increasingly adopted in teleconsultation for its convenience and mobility. Although widely accepted by physicians for informal online consultations, the effectiveness of mobile platforms in formal medical interventions, such as follow-up treatments, remains largely underexplored. This research presents a case study from a Chinese hospital, examining how physicians use mobile platforms to treat chronic patients, the challenges they encounter, and their overall experience. The study aims to determine whether mobile technology can enable physicians to fulfill their responsibilities in managing chronic disease follow-ups online. Through observations and interviews, we found that using mobile platforms introduces significant challenges. Physicians operate in complex and varying environments, such as workplaces, homes, and even public areas, making them struggle to access comprehensive information, communicate effectively, and maintain detailed medical records. As a result, physicians face poor working conditions, reduced efficiency, and struggle to make accurate treatment decisions. This study highlights the need for policy reforms and technological innovations to ensure sustainable teleconsultation practices on mobile platforms. It contributes to the HCI and CSCW communities by highlighting the feasibility of using mobile platforms in online chronic disease follow-ups.

Low-resource relation extraction (LRE) aims to extract the relationships between given entities from natural language sentences in low-resource application scenarios, which has been an incredibly challenging task due to the limited annotated corpora. Existing studies either leverage self-training schemes to expand the scale of labeled data, while the error accumulation of pseudo-labels' selection bias provoke the gradual drift problem in subsequent relation prediction, or utilize the instance-wise contrastive learning that fails to distinguish those sentence pairs with similar semantics. To alleviate these defects, this article introduces a novel contrastive learning framework called hierarchical relation contrastive learning (HRCL) for LRE. HRCL leverages task-related instruction description and schema-constrained as prompts to generate high-level relation representations. To enhance the efficacy of contrastive learning, we further employ hierarchical affinity propagation clustering (HiPC) to derive hierarchical signals from relational feature space with a hierarchy cross-attention (HCA) mechanism and effectively optimize pair-level relation features through relation-wise contrastive learning. Exhaustive experiments have been conducted on five public relation extraction (RE) datasets in low-resource settings. The results demonstrate the effectiveness and robustness of HRCL and outperform the current state-of-the-art (SOTA) model by 6.56% on average in terms of B3F1. Our source code is publicly available at https://github.com/Phevos75/HRCLRE.

In hyperspectral image (HSI) classification, the challenge of the small-sample-size problem persists as a significant obstacle due to the high cost of labeling samples. To effectively train models with a limited sample set, the application of a transfer learning approach called cross-scene HSI classification is considered a viable solution to address this problem. In cross-scene HSI classification, a source scene with sufficient labeled samples is leveraged to assist in classifying a target scene that lacks labeled samples. Considering that real HSIs may be captured by different sensors, we propose a novel heterogeneous transfer learning algorithm called dual-stream discriminative attention network (DSDAN) to address the task of cross-scene HSI classification. The DSDAN predominantly comprises three pivotal modules. 1) A dual-stream lightweight hybrid CNN (DSLHC) incorporates both the source stream and the target stream is applied to extract alignment spatial-spectral features from heterogeneous data. 2) A discriminative attention block (DAB) is created to address the domain shift between two scenes. Following the DSLHC, the DAB assigns discriminative attention weights to the source features, facilitating a closer alignment of features from two scenes. 3) A specially designed cross-domain loss (CDL) is designed to drive intra-class samples from two scenes to become more consistent, while inter-class samples from two scenes become more distinct, thereby further mitigating domain shift. By combining DSLHC, DAB, and CDL, the complete DSDAN model is established. The effectiveness of DSDAN is validated using three real cross-scene HSI datasets.

Temporal Knowledge Graph Question Answering (TKGQA) task aims to find an entity or timestamp from a temporal knowledge graph to answer temporal reasoning questions. However, most existing models fail to capture the implicit temporal information in the questions, resulting in weak performance when handling complex temporal reasoning tasks. To address this issue, this paper proposes a novel TKGQA model called GATQR, which integrates graph attention mechanism. The model utilizes a pre-trained temporal knowledge base in the form of quadruples and introduces Graph Attention Network (GAT) to effectively capture the implicit temporal information in the questions. By integrating with relation representations trained by the RoBERTa, it further enhances the temporal relationship representation in the queries. Finally, this representation is combined with the pre-trained TKG embeddings to predict the entity or timestamp with the highest score as the answer. Experimental results on the largest benchmark dataset CronQuestion demonstrate that compared to baseline models such as CronKGQA, EntityQR, and TempoQR-Soft, the GATQR achieves significant improvements in Hits@l results for handling complex and temporal question types, with increases of 35% and 13%, 18% and 9%, and 9% and 3%, respectively. These results validate the effectiveness and superiority of the GATQR model in capturing implicit temporal information and enhancing complex reasoning capabilities.

Vision Language Pre-training (VLP) has made significant progress in the field of universal multimodality in recent years. Universal multimodal datasets (such as MSCOCO, Flickr30k, etc.) have become one of the standards for evaluating VLP models which rely on images and corresponding captions for representation modeling. However, in a public opinion event, besides image captions, it also includes other texts such as content texts and comments, which may have a positive impact on image-text representations of public opinion. In this paper, we propose the method to explore the positive effect of content texts on initial multimodal representations. We name our model CLIP-PubOp, which is based on CLIP, a famous VLP model using contrastive learning. We add a linear fusion layer before the fusion of image and text representations, which fuses event content text representations and caption representations in a certain proportion to obtain enhanced text representations. On this basis, multimodal representations can be obtained by fusing enhanced text representations with image representations. We crawl through four mainstream categories of public opinion events online as our datasets and conduct experiments on both Chinese and English version of datasets. The experimental results show that content texts of public opinion events have a significant positive effect on multimodal representations, with an average accuracy improvement of about 2%-10% in image-text retrieval tasks.

Generative Adversarial Networks (GANs) are capable of effectively synthesising new realistic images and estimating the potential distribution of samples utilising adversarial learning. Nevertheless, conventional GANs require a large amount of training data samples to produce plausible results. Inspired by the capacity for humans to quickly learn new concepts from a small number of examples, several meta-learning approaches for the few-shot datasets are presented. However, most of meta-learning algorithms are designed to tackle few-shot classification and reinforcement learning tasks. Moreover, the existing meta-learning models for image generation are complex, thereby affecting the length of training time required. Fast Adaptive Meta-Learning (FAML) based on GAN and the encoder network is proposed in this study for few-shot image generation. This model demonstrates the capability to generate new realistic images from previously unseen target classes with only a small number of examples required. With 10 times faster convergence, FAML requires only one-fourth of the trainable parameters in comparison baseline models by training a simpler network with conditional feature vectors from the encoder, while increasing the number of generator iterations. The visualisation results are demonstrated in the paper. This model is able to improve few-shot image generation with the lowest FID score, highest IS, and comparable LPIPS to MNIST, Omniglot, VGG-Faces, andminiImageNet datasets. The source code is available onhttps://github.com/phaphuang/FAML.

With the exponential growth of video data, video summarization techniques are urgently needed for reducing people’s efforts in the videos' content exploration by generating succinct but informative summaries from original lengthy videos. Though supervised video summarization approaches have demonstrated the state-of-the-art performance, unsupervised methods are still highly demanded due to resourcefully expensive human annotations and the subjectiveness of video summarization tasks. In this paper, a novel unsupervised-based Deep Self-attention Recurrent summarization network with Reinforcement Learning (DSR-RL) for video summarization is proposed. The model can learn the input video sequence and suggest the key-shot summary without additional human annotations by integrating self-attention, BRNN, and reinforcement learning mechanisms. The DSR-RL improves not only importance score through the attention map vector of self-attention network but also the diversity of summaries via the reward function of reinforcement learning. Our method outperforms the state-of-the-art unsupervised video summarization methods on both SumMe and TVSum datasets. The source code is available at https://github.com/phaphuang/DSR-RL.

The purpose of this study is to conduct a bibliometric analysis to examine the most influential journals, institutions, and countries in social media (SM) publications related to knowledge management (KM). Moreover, various research themes in SM KM publications are also explored. VOSviewer was employed to process 234 SM KM publications retrieved from Web of Science (WoS) in the time period 2009-2019. Different methodologies were used according to the nature of bibliometric analysis and explained in each section. Journal of Knowledge Management was the most influential journal in SM KM publications. USA and England ranked first and second respectively, while the Tampere University of Technology was the most productive institute in SM KM research. Four emerged themes indicated an explicit contribution of SM users in KM through big data, knowledge sharing, innovation, Enterprise 2.0, and social capital. This is the first bibliometric study that explores the overall contribution of SM publications in the KM field.

In literature, there is a shortage of comprehensive documents that can provide proper details about Twitter in research community. This study conducted a first descriptive bibliometric analysis to examine the most influential journals, institutions, and countries on Twitter. Similarly, bibliometric mapping analysis is carried out to explore different research themes in Twitter publications. VOSviewer was employed to process the 11,006 Twitter publications retrieved from the Web of Science (WoS) from 2009 to 2018. Obtained results suggest that USA and China received the highest number of publications on Twitter research, while the University of Illinois was the most productive institute. Furthermore, the five major themes have emerged in Twitter publications, and its remarkable role has been found in event detection, sentiment analysis, education, health, politics, and crisis as well as risk management. The authors believe that this study will open new doors for researchers to use online Twitter social networking communities in beauty salons, consulting companies, banks, and airlines.

Sentiment analysis is significant for excavating text opinion. There are two issues in the foreign exchange (Forex) field. 1) In sentiment orientation, most researches focus on product reviews, lack fine-grained sentiment analysis for Forex news. 2) In sentiment intensity, most works consider the intensity of sentiment words but ignore the significance of field characteristics. Aiming at the two problems, a fine-grained Sentiment Analysis model (shorted as WD-SA) is established, which integrates with the Weight of sentiment words and Domain features. First, the semantic information of text is embedded into a vector based on word2vec. Then, sentiment orientation is detected by a method, which combines machine learning algorithm and the weight of sentiment words. Finally, features are extracted to investigate the intensity of news. The experimental results show that our algorithm outperforms the state-of-the-art.

The forecast and analysis of the trend of the RMB exchange rate have been deeply explored by many researchers in the financial field, but the combination of public opinion sentiment data and historical market data to forecast the RMB exchange rate in the short term has been studied little. With the rapid development of the Internet, the influence of public opinion sentiment on the economy and society is increasing. Online public opinion sentiment data not only have an impact on stock prices [1] and commodity prices, but also have a significant impact on foreign exchange (Forex) rates. However, the public opinion data are not applied for the RMB exchange rate forecast, because the impact of public events on the exchange rate is ignored. Besides, the lack of exact temporal sliding window of public opinion ignores its timeliness and sensibility.

The convenience of online social media in communication and information dissemination has made it an ideal place for spreading rumor events and automatically debunking rumor events is a crucial problem. However, it is a challenging task to employ traditional classification approaches to rumor events detection since they rely on hand-crafted features which require daunting manual efforts. Besides, the various posts on a rumor event will debate its realness over time, and the distribution of the posts is special in time dimension. Thus, this paper presents a novel method for rumor event detection based on a dynamic time series (DTS) algorithm and a two layer Gated Recurrent Unit (GRU) model, named 2-GRU-DTS. The proposed model uses the DTS algorithm to retain the distribution information of social events over time and uses the two layers GRU model to learn the hidden event representations. Experimental results on real datasets from Sina Weibo demonstrate that our proposed 2-GRU-DTS model outperforms latest rumor event detection algorithms.

Patent, an important scientific literature, records a large amount of innovative and practical research.The patent keywords also provide a high-level topic description of a patent document and hold an important position in classic NLP tasks, such as patent classification or clustering.However, there are few research works on keywords extraction covering the Chinese patents in current stage.In this paper, we propose a novel algorithm to extract keywords from Chinese patents.A sentenceranking model, based on a sentence embedding graph and heuristic rules, is constructed to select the top-KS percent of the sentences.At the same time, the semantic-ranking weights of sentences are also transmitted to keywords extraction.The experimental results on our Chinese patents datasets testifies that the sentence-ranking based keywords extraction algorithm improves the performance by 6% to 13% in F-score.In summary, the new idea of selecting key sentences from original documents can effectively filter out noisy sentences and leverage the performance of keywords extraction.