卷 23, 编号 5 (2024)

Chatbots have become interesting for many users as technology becomes more and more advanced. The need for information exchange among people through computer systems is increasing daily, raising the preference for using chatbots in most countries. Since Vietnam is such a developing country with a variety of ethnic groups, it requires much attention to the proliferation of social networks and the expansion of the cooperative economy. Regarding social networks, the inappropriate use of words in everyday life has become a significant issue. There are mixed reviews of praise and criticism on social networks; and we try to reduce the negative language use and improve the quality of using social networks language. We aim to meet users’ needs on social networks, promote economic development, and address social issues more effectively. To achieve these goals, in this paper we propose a deep learning technique using ontology knowledge mining to collect and process comments on social networks. This approach aims to enhance the user experience and facilitate the exchange of information among people by mining opinions in comments. Experimental results demonstrate that our method outperforms the conventional approach.

A network is an aggregation of nodes joined by edges, representing entities and their relationships. In social network clustering, nodes are organized into clusters according to their connectivity patterns, with the goal of community detection. The detection of community structures in networks is essential. However, existing techniques for community detection have not yet utilized the potential of the Fish School Search (FSS) algorithm and modularity principles. We have proposed a novel method, clustering with the Fish School Search algorithm and modularity function (FSC), that enhances modularity in network clustering by iteratively partitioning the network and optimizing the modularity function using the Fish School Search Algorithm. This approach facilitates the discovery of highly modular community structures, improving the resolution and effectiveness of network clustering. We tested FSC on well-known and unknown network structures. Also, we tested it on a network generated using the LFR model to test its performance on networks with different community structures. Our methodology demonstrates strong performance in identifying community structures, indicating its effectiveness in capturing cohesive communities and accurately identifying actual community structures.

IoT devices are crucial in this modern world in many ways, as they provide support for environmental sensing, automation, and responsible resource conservation. The immense presence of IoT devices in everyday life is inevitable in the smart world. The predominant usage of IoT devices lurks the prying eyes of intentional hackers. Though there are several precautionary security systems and protocols available for generic wireless networks, it is observed that there is a need to formulate a state-of-the-art security mechanism exclusively for IoT network environments. This work is submitted here for the betterment of IoT network security. Three dedicated contributions are integrated in this work to achieve higher security scores in IoT network environments. Fast Fuzzy Anomaly Detector, Legacy Naïve Bayes Attack Classifiers, and Variable Security Schemer of Rivest-Shamir-Adleman algorithm are the novel modules introduced in this work abbreviated as ASORI. Captivating the advantages of the onboard IoT certification mechanism and selecting a dynamic security strategy are the novelties introduced in this work. ASORI model is tested with industrial standard network simulator OPNET to ensure the improved security along with vital network performance parameter betterments.

The paper proposes a method for using Fuzzy controller tools to synthesize optimal control of a second-order dynamic object with a delay. The idea is to construct a phase surface that combines optimal relay control away from the equilibrium state region and linear control in the region itself. This approach made it possible to avoid self-oscillations in a steady state while maintaining the properties of optimal control in terms of speed. The switching trajectory in phase space, corresponding to the solution of the optimal control problem according to the Maximum principle, is determined by the method of inverse time calculation of the second-order difference equation of the plant. The region near the equilibrium state, where the linear controller is used, is determined from the results of modeling the motion of a point in phase space with optimal control for an object with a delay. This region is represented by an ellipse that describes motion in phase space in a self-oscillatory mode. To further eliminate self-oscillations in this area, a linear controller is used, tuned by means of solving the variational optimal control problem. It is proposed to use a tool for the synthesis of Fuzzy controllers, where the switching surface and calculation of the control value can be set arbitrarily. As a result, a variable controller structure is proposed to combine these two approaches. The resulting Fuzzy controller model is represented by a standard FLS structure, which was implemented in Python on the Orange Pi embedded computer. To connect to the current control object, an industrial controller FX3U-24MR is used, connected to a computer via a Modbus network. Full-scale tests are presented on a hot water supply temperature control object, which corresponds as closely as possible to the object model under study. The method, idea and results obtained in the work can be applied and investigated in the synthesis of control of dynamic objects in sliding mode to solve current problems related to eliminating the unwanted Chattering effect.

Orientation and positioning control of an unmanned aerial vehicle (UAV) vertical take-off and landing multi-rotor type in space is inextricably linked with the formation of a motion control vector, consisting of a combination of thrusts and aerodynamic moments of each propeller-motor group. The accuracy and speed of formation of the motion control vector greatly affect the positioning and orientation errors of the UAV. Most works devoted to the synthesis of UAV control systems use a motion control vector without taking into account the dynamics of the rotor-motor groups, which in some cases forces the control system to reduce its performance. The performance of the UAV control system can be increased by increasing the speed of generation of the thrust of the propeller-motor groups, for which a system for controlling the thrust of the propeller-motor group has been proposed. The propeller-motor group in its composition has a nonlinear internal connection in the aerodynamic torque and an output signal – thrust, that nonlinearly depends on the square of the propeller rotation speed. Typically, the propeller group is controlled like an electric motor – the internal coupling of the aerodynamic torque is considered an external disturbance, and the thrust is controlled by changing the speed of rotation of the propeller which is calculated based on the required motion control vector. It is proposed to consider thrust and aerodynamic torque an integral part of the propeller-engine group, for which to build a linear thrust control system. For this purpose, we carried out feedback linearization of the rotor-motor group system, connecting the voltage supplied to the motors with the motion control vector, which is the output value. The linearization process is divided into two stages: at the first stage feedback linearization is performed for an electric motor with internal nonlinear coupling by aerodynamic torque; at the second stage, linearization is performed with feedback on the output obtained at the first stage of the system with a nonlinear output signal – thrust. In accordance with the principles of subordinate control, motor control is formed for linearized feedback of the propeller group. Simulation was completed. An important issue when using feedback linearization is the preservation of the quality characteristics of the control system in the event of a mismatch between the parameters of the object and the model, the parameters of which are used to calculate the linearizing feedback. In this work, modeling was carried out with a discrepancy of some parameters up to 50%.