Intelligent computation offloading and resource allocation for mobile users (MUs) in ubiquitous edge Internet of Things (IoT) systems is a worthy research hotspot. To improve the latency and energy consumption of MUs in ubiquitous edge IoT systems, we propose a deep reinforcement learnin (DRL)-driven intelligent joint optimization strategy for computation offloading and resource allocation that includes relay selection, offloading decisions, and resource allocation. Specifically, according to the limited coverage of base stations and the extremely high deployment costs in actual environments, we introduce the virtual backbone architecture to provide users with efficient multi-hop offload services through a connected dominating set (CDS). Then, we propose a CDS-based deep reinforcement learning algorithm to search for the shortest path from the MUs to the multi-access edge computing (MEC) server. Furthermore, based on the highly coupled relationship between the offloading decision and resource allocation, we design a DLIO algorithm to solve for the joint optimization of computation offloading and resource allocation. The experimental results demonstrate that our proposed optimization algorithm outperforms the state-of-the-art methods in terms of total system cost, success rate, and acceptance number.

Document clustering is of high importance for many natural language technologies. A wide range of computational traditional topic models, such as LDA (Latent Dirichlet Allocation) and its variants, have made great progress. However, traditional LDA-based clustering algorithms might not give good results due to such probabilistic models require prior distributions which are always difficult to define. In this paper, we propose a probabilistic model named tpLDA, which incorporates different levels of topic popularity information to determine the prior LDA distribution, discover the latent topics and achieve better clustering. Specifically, global topic popularity is introduced to reduce the potential distraction in local cluster popularity and the local cluster popularity draws more attention on certain parts of the global topic popularity. The two popularities contribute complementary information and their integration can dynamically adjust statistical parameters of the model. Experimental evaluations on real data sets show that, compared with state-of-the-art approaches, our proposed framework dramatically improves the accuracy of documents clustering.

Stance detection on social media aims to identify the stance of social media users toward a topic or claim, which can provide powerful information for various downstream tasks. Many existing stance detection approaches neglect to model the deep semantic representation information in tweets and do not explore aggregating the hierarchical features among words, thus degrading performance. To address these issues, this article proposes a novel deep learning approach P retrained E mbeddings for Stance Detection with H ierarchical C apsule N etwork (PE-HCN) without complicated preprocessing. Specifically, PE-HCN first adopts a pretrained language model and then uses a related textual entailment task for fine-tuning to obtain the deep textual representations of tweets. The PE-HCN approach extends the dynamic routing scheme to cope with these deep textual representations by utilizing primary capsules for routing the information among words in each tweet and applying secondary capsules to transmit the aggregated features to each category capsule accordingly. Moreover, to improve the confidences of the category capsules, we design an adaptive feedback mechanism to dynamically strengthen the routing signals. Through experiments on three benchmark datasets, compared with the state-of-the-art baselines, the extensive results exhibit that PE-HCN achieves competitive improvements of up to 6.32%, 2.09%, and 1.8%, respectively.

Topic models have been widely utilized in Topic Detection and Tracking tasks, which aim to detect, track, and describe topics from a stream of broadcast news reports. However, most existing topic models neglect semantic or syntactic information and lack readable topic descriptions. To exploit semantic and syntactic information, Language Models (LMs) have been applied in many supervised NLP tasks. However, there are still no extensions of LMs for unsupervised topic clustering. Moreover, it is difficult to employ general LMs (e.g., BERT) to produce readable topic summaries due to the mismatch between the pretraining method and the summarization task. In this paper, noticing the similarity between content and summary, first we propose a Language Model-based Topic Model (LMTM) for Topic Clustering by using an LM to generate a deep contextualized word representation. Then, a new method of training a Topic Summarization Model is introduced, where it is not only able to produce brief topic summaries but also used as an LM in LMTM for topic clustering. Empirical evaluations of two different datasets show that the proposed LMTM method achieves better performance over four baselines for JC, FMI, precision, recall and F1-score. Additionally, the generated readable and reasonable summaries also validate the rationality of our model components.