Transfer learning has achieved promising results by leveraging knowledge from the source domain to annotate target which few or none labels. Existing methods often seek minimize distribution divergence between domains, such as marginal distribution, conditional both. However, these two distances are treated equally in existing algorithms, will result poor performance real applications. Moreover, usually assume that dataset is balanced, also limits their performances on imbalanced tasks quite...

Trust is an important concept in human interactions which facilitates the formation and continued existence of functional societies. In first decade 21st century, computational trust models have been applied to solve many problems wireless communication systems. This cross-disciplinary research has yielded innovative solutions. this paper, we examine latest methods proposed by researchers manage reputation Specifically, survey state art application fields mobile ad hoc networks (MANETs),...

1Food safety is becoming more and serious topic worldwide. To tackle the food issues from technical aspect, people need a trusted traceability system that can track monitor whole lifespan of production, including processes raw material cultivation/breeding, processing, transporting, warehousing, selling etc. In this paper, we propose trusted, self-organized, open ecological based on blockchain Internet Things (IoT) technologies, which involves all parties smart agriculture ecosystem, even if...

In open and dynamic multiagent systems (MASs), agents often need to rely on resources or services provided by other accomplish their goals. During this process, are exposed the risk of being exploited others. These risks, if not mitigated, can cause serious breakdowns in operation MASs threaten long-term wellbeing. To protect from uncertainty behavior interaction partners, age-old mechanism trust between human beings is re-contexted into MASs. The basic idea let self-police MAS rating each...

As artificial intelligence (AI) systems become increasingly ubiquitous, the topic of AI governance for ethical decision-making by has captured public imagination. Within research community, this remains less familiar to many researchers. In paper, we complement existing surveys, which largely focused on psychological, social and legal discussions topic, with an analysis recent advances in technical solutions governance. By reviewing publications leading conferences including AAAI, AAMAS,...

Multimodal recommender systems utilizing multimodal features (e.g., images and textual descriptions) typically show better recommendation accuracy than general models based solely on user-item interactions. Generally, prior work fuses into item ID embeddings to enrich representations, thus failing capture the latent semantic item-item structures. In this context, LATTICE proposes learn structure between items explicitly achieves state-of-the-art performance for recommendations. However, we...

Activity recognition using mobile phones has great potential in many applications including healthcare. In order to let a person easily know whether he is strict compliance with the doctor's exercise prescription and adjust his amount accordingly, we can use smart-phone based activity reporting system accurately recognize range of daily activities report duration each activity. A triaxial accelerometer embedded smart phone used for classification several activities, such as staying still,...

With1 the rapid growth of internet things (IoT) market and requirement, low power wide area (LPWA) technologies have become popular. In various LPWA technologies, Narrow Band IoT (NB-IoT) long range (LoRa) are two main leading competitive technologies. Comparing to NB-IoT network that mainly built managed by mobile operators, LoRa wide-area (LoRaWAN) is operated private companies or organizations, which will bring trust issues between application customers operations. this paper, we proposed...

Cognitive radio (CR) is a promising concept for improving the utilization of scarce spectrum resources. A reliable strategy detection unused bands essential to design and practical implementation CR systems. It widely accepted that in real-world environment, cooperative sensing involving many secondary users scattered wide geographical area can greatly improve accuracy. However, some may misbehave, i.e. provide false information, an attempt maximize their own utility gains. Such selfish...

Learning semantic-rich representations from raw unlabeled time series data is critical for downstream tasks such as classification and forecasting. Contrastive learning has recently shown its promising representation capability in the absence of expert annotations. However, existing contrastive approaches generally treat each instance independently, which leads to false negative pairs that share same semantics. To tackle this problem, we propose MHCCL, a Masked Hierarchical Cluster-wise...

Next Point-Of-Interest (POI) recommendation plays an important role in various location-based services. Its main objective is to predict the user's next interested POI based on her previous check-in information. Most existing methods directly use users' historical trajectories construct graphs assist sequential models complete this task. However, as data extremely sparse, it difficult capture potential relations between POIs by using these data. To end, we propose Sequence-based Neighbour...

In contrast to traditional recommender systems which usually pay attention users' general and long-term preferences, sequential recommendation (SR) can model dynamic intents based on their behaviour sequences suggest the next item(s) them. However, most of existing models learn ranking score an item only its relevance property, personalized user demands in terms different learning objectives, such as diversity, tail novelty or recency, have been played essential roles multi-objective (MOR),...

Fuzzy cognitive maps (FCMs), as an illustrative causative representation of modeling and manipulation complex systems, can be used to model the dynamic behavior investigated systems. However, due defects in expression architecture, traditional FCMs most their relevant extensions are not applicable classification problems. To solve this problem, paper presents approach that directly extends by translating reasoning mechanism a set fuzzy if– then rules. Moreover, proposed fully considers...

Cognitive maps (CMs), fuzzy cognitive (FCMs), and dynamical networks (DCNs) are related tools for modeling the cognition of human beings facilitating machine inferences accordingly. FCMs extend CMs, DCNs FCMs. Domain experts often face challenge that CMs/FCMs not sufficiently capable in many applications too complex. This paper presents a simplified DCN (sDCN) extends capability FCM/CM, yet maintains simplicity. Additionally, this proves there exists theoretical equivalence among models map...

Crowd sourcing (CS) systems offer a new way for businesses and individuals to leverage on the power of mass collaboration accomplish complex tasks in divide-and-conquer manner. In existing CS systems, no facility has been provided analyzing trustworthiness workers providing decision support allocating workers, which leads high dependency quality work behavior as shown this paper. To address problem, trust management mechanisms are urgently needed. Traditional techniques focused identifying...

Purpose With the rapid growth of Internet Things (IoT) market and requirement, low power wide area (LPWA) technologies have become popular. In various LPWA technologies, Narrow Band IoT (NB-IoT) long range (LoRa) are two main leading competitive technologies. Compared with NB-IoT networks, which mainly built managed by mobile network operators, LoRa networks (LoRaWAN) operated private companies or organizations, suggests issues: trust operators lack coverage. This study aims to propose a...

From sophisticated single agent in complex environments to multi-agent system (MAS) organizations, intelligent software research has come a long way just under two decades. Many new branches of this field have emerged over the years which enabled today's agents perform wide variety human-like tasks such as learning, reasoning, negotiating, self-organizing and trusting each other, etc. Unfortunately, very few practical MASs been deployed after period intensive development. For achieve higher...

Vehicle tracking detection, recognition and counting is an important part of vehicle analysis. Designing such a model with excellent performance difficult. The traditional target detection algorithm based on artificial features has poor generalization ability robustness. In order to take use the deep learning method for counting, this paper proposes yolov5. This uses technology, takes running vehicles video as research object, analysis algorithm, framework platform. relevant platform...

Mining potential and valuable medical knowledge from massive data to support clinical decision-making has become an important research field. Personalized medicine recommendation is direction in this field, aiming recommend the most suitable medicines for each patient according health status of patient. can assist clinicians make decisions avoid occurrence abnormalities, so it been widely concerned by many researchers. Based on this, paper makes a comprehensive review personalized...

Sequential recommendation systems often suffer from data sparsity, leading to suboptimal performance. While multimodal content, such as images and text, has been utilized mitigate this issue, its integration within sequential frameworks remains challenging. Current models are unable effectively explore capture correlations among behavior sequences of users items across different modalities, either neglecting sequence representations or inadequately capturing associations between in their...