In crowd evacuation research, the knowledge contained in crowd evacuation is very complex and is multisource. Crowd evacuation scenarios restrict pedestrians’ movement decision-making, and the movement states of the crowd imply the movement characteristics. However, the existing studies on crowd evacuation navigation approach cannot make full use of the complex and multisource crowd evacuation knowledge, which reduces the effect of the evacuation navigation. To solve this problem, a new crowd evacuation navigation approach based on multisource knowledge is proposed. First, we collect relevant data on crowd evacuation using an image sensor network and establish a crowd evacuation knowledge graph to organize and store this data. Second, the explicit knowledge of scene structure and crowd movements is represented based on the crowd evacuation knowledge graph. Then, a deep-learning-based tacit knowledge model (DLTKM) is designed to extract the tacit knowledge of different groups and scene entities. Finally, a new crowd evacuation navigation approach based on wireless sensor network and related knowledge representations is designed to plan evacuation paths for evacuees. The experiment results show that this approach can plan reasonable evacuation paths for pedestrians, and improve the efficiency of crowd evacuations.

Nowadays recording and sharing personal lives using mobile devices on the Internet is becoming increasingly popular, and successive POI recommendation is gaining growing attention from academia and industry. In mobile scenarios, multiple influencing factors including the diversity of user preferences, the changeability of user behavior and the dynamic of spatiotemporal context bring great challenges to the POI recommender system. In order to accurately capture both the stable and the contextual preferences of mobile users in dynamic contexts, we propose a fusion framework JANICP (Joint Attention Networks with Inherent and Contextual Preferences) for successive POI recommendation by jointly training an offline/nearline user inherent interest perception model and an online user contextual interest prediction model. The offline model is trained based on the global historical behavior data to achieve stable interest representation, while the online model is trained based on the instantly selected context-sensitive data to achieve dynamic interest perception. An attention aggregation and matching module is used to fully connect the two kinds of preference representations and generate the final POI recommendation. Extensive experiments were conducted on three real datasets and experimental results show that the proposed JANICP outperforms existing state-of-the-art methods.

Nowadays, data centers are critical infrastructure for the information industry. Thermal security is one of the most concerning problems of the data center efficiently providing service. The temperature prediction method is an effective way, which overcomes the lagging of the feedback control and rewards a high prediction accuracy. While the current LSTM based prediction methods are limited in accuracy and restricted in scalability due to the lack of knowledge of physical properties and consideration of time constant differences of features. To address this, we propose a data center temperature prediction model with two-segment LSTM for prediction separately for IT equipment load and other heat-related variables with different time constants. The model takes into account the physical properties of the equipment and achieves higher prediction accuracy. The experimental results show that the prediction accuracy of our method is 27.27% higher than the state-of-art single segment LSTM method.

Human Activity Recognition (HAR) enables computer systems to assist users with their tasks and improve their quality of life in rehabilitation, daily life tracking, fitness, and cognitive disorder therapy. It is a hot topic in the field of machine learning, and HAR is gaining more attention among researchers due to its unique societal and economic advantages. This paper focuses on a collaborative computation scenario where a group of participants will securely and collaboratively train an accurate HAR model. The training process requires collecting a massive number of personal activity features and labels, which raises privacy problems. We decentralize the training process locally to each client in order to ensure the privacy of training data. Furthermore, we use an advanced secure aggregation algorithm to ensure that malicious participants cannot extract private information from the updated parameters even during the aggregation phase. Edge computing nodes have been introduced into our system to address the problem of data generation devices’ insufficient computing power. We replace the traditional central server with smart contract to make the system more robust and secure. We achieve the verifiability of the packaged nodes using the publicly auditability feature of blockchain. According to the experimental data, the accuracy of the HAR model trained by our proposed framework reaches 93.24%, which meets the applicability requirements. The use of secure multiparty computation techniques unavoidably increases training time, and experimental results show that a round of iterations takes 36.4 seconds to execute, which is still acceptable.

Referring Expression Comprehension (REC) is a multimodal comprehension task that aims to locate an object in an image, given a text description. Traditionally, during the existing REC tasks, there has been a basic assumption that the given text expression and the image are usually exactly matched to each other. However, in real-world scenarios, there is uncertainty in how well the image and text match each other exactly. Illegible objects in the image or ambiguous phrases in the text have the potential to significantly degrade the performance of conventional REC tasks. To overcome these limitations, we consider a more practical and comprehensive REC task, where the given image and its referring text expression can be inexactly matched. Our models aim to correct such inexact matching and supply corresponding interpretations. We refer to this task as Further REC (FREC) . This task is divided into three subtasks: 1) correcting the erroneous text expression using visual information, 2) generating the rationale for this input expression, and 3) localizing the proper object based on the corrected expression. We introduce three new datasets for FREC: Further-RefCOCOs , Further-Copsref and Further-Talk2Car . These datasets are based on the existing REC datasets, including RefCOCO and Talk2Car. We developed a novel pipeline architecture to execute the three subtasks simultaneously in an end-to-end fashion. Next, we developed an elastic masked language modeling (EMLM) training head to rectify text errors with uncertain lengths. Our experimental results demonstrate the validity of our proposed pipeline. We hope this work sparks more research focused on inexactly matched REC.