With the increased popularity of mobile devices and smart phones, location-based services (LBS) have become a common need in our daily life. Therefore, maintaining the correctness of POI (Points of Interest) data has become an important issue for many location-based services such as Google Maps and Garmin navigation systems. The simplest form of POI contains a location (e.g., represented by an address) and an identifier (e.g., an organization name) that describes the location. As time goes by, the POI relationship of a location and organization pair may change due to the opening, moving, or closing of a business. Thus, effectively identifying outdated or emerging POI relations is an important issue for improving the quality of POI data. In this paper, we examine the possibility of using location-related pages on the Web to verify existing POI relations via weakly labeled data, e.g., the co-occurrence of an organization and an address in Web pages, the published date of such pages, and the pairing diversity of an address or an organization, etc. The preliminary result shows a promising direction for discovering emerging POI and mandates more research for outdated POI.
With the rise of artificial intelligence, conversational agents (CA) have found use in various applications in the commerce and service industries. In recent years, many conversational datasets have becomes publicly available, most relating to open-domain social conversations. However, it is difficult to obtain domain-specific or language-specific conversational datasets. This work focused on developing conversational systems based on the Chinese corpus over military scenarios. The soldier will need information regarding their surroundings and orders to carry out their mission in an unfamiliar environment. Additionally, using a conversational military agent will help soldiers obtain immediate and relevant responses while reducing labor and cost requirements when performing repetitive tasks. This paper proposes a system architecture for conversational military agents based on natural language understanding (NLU) and natural language generation (NLG). The NLU phase comprises two tasks: intent detection and slot filling. Detecting intent and filling slots involves predicting the user’s intent and extracting related entities. The goal of the NLG phase, in contrast, is to provide answers or ask questions to clarify the user’s needs. In this study, the military training task was when soldiers sought information via a conversational agent during the mission. In summary, we provide a practical approach to enabling conversational agents over military scenarios. Additionally, the proposed conversational system can be trained by other datasets for future application domains.
Recent developments have made software-defined networking (SDN) a popular technology for solving the inherent problems of conventional distributed networks. The key benefit of SDN is the decoupling between the control plane and the data plane, which makes the network more flexible and easier to manage. SDN is a new generation network architecture; however, its configuration settings are centralized, making it vulnerable to hackers. Our study investigated the feasibility of applying artificial intelligence technology to detect abnormal attacks in an SDN environment based on the current unit network architecture; therefore, the concept of symmetry includes the sustainability of SDN applications and robust performance of machine learning (ML) models in the case of various malicious attacks. In this study, we focus on the early detection of abnormal attacks in an SDN environment. On detection of malicious traffic in SDN topology, the AI module in the topology is applied to detect and act against the attack source through machine learning algorithms, making the network architecture more flexible. Under multiple abnormal attacks, we propose a hierarchical multi-class (HMC) architecture to effectively address the imbalanced dataset problem and improve the performance of minority classes. The experimental results show that the decision tree, random forest, bagging, AdaBoost, and deep learning models exhibit the best performance for distributed denial-of-service (DDoS) attacks. In addition, for the imbalanced dataset problem of multiclass classification, our proposed HMC architecture performs better than previous single classifiers. We also simulated the SDN topology and scenario verification. In summary, we concatenated the AI module to enhance the security and effectiveness of SDN networks in a practical manner.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.