A Conversation History-Based Q&A Cache Mechanism for Multi-Layered Chatbot Services

Chatbot technologies have made our lives easier. To create a chatbot with high intelligence, a significant amount of knowledge processing is required. However, this can slow down the reaction time; hence, a mechanism to enable a quick response is needed. This paper proposes a cache mechanism to improve the response time of the chatbot service; while the cache in CPU utilizes the locality of references within binary code executions, our cache mechanism for chatbots uses the frequency and relevance information which potentially exists within the set of Q&A pairs. The proposed idea is to enable the broker in a multi-layered structure to analyze and store the keyword-wise relevance of the set of Q&A pairs from chatbots. In addition, the cache mechanism accumulates the frequency of the input questions by monitoring the conversation history. When a cache miss occurs, the broker selects a chatbot according to the frequency and relevance, and then delivers the query to the selected chatbot to obtain a response for answer. This mechanism showed a significant increase in the cache hit ratio as well as an improvement in the average response time.

Designing RESTful API for the e-procurement system in private sector

The software for the e-procurement system was developed based on .NET Core RESTful API with Open API specifications. The server side uses RESTful API which ensures compatibility with the ma-jority of clients and enables them to exchange information in JSON format. The authentication and authorization flow was implemented using OAuth open standard paired with Microsoft Identity Service. User roles and functionality were handled with a standalone service for authentication and registration that made our system efficient and scalable. Business logic was designed to be split into micro-services accessible through rout-ing controllers. This approach allowed us to separate the responsibilities between the server and the client side. Special authorization headers passed during modi-fication queries allowed us to control and restrict access to particular resources for unauthorized users. The distributed cache mechanism inside the data repository level was used in order to increase the responsiveness of the system. The state handling subsystem was designed utilizing Finite State Machine concepts. The developed system was verified using unit and integration tests.

Distributed Top-K Join Queries Optimizing for RDF Datasets

In recent years, the scale of RDF datasets is increasing rapidly, the query research on RDF datasets in the transitional centralized environment is unable to meet the increasing demand of data query field, especially the top-k query. Based on the Spark distributed computing system and the HBase distributed storage system, a novel method is proposed for top-k query. A top–k query plan STA (Spark Threshold Algorithm) is proposed to reduce the connection operation of RDF data. Furthermore, a better algorithm SSJA (Spark Simple Join Algorithm) is presented to reduce the sorting related operations for the intermediate data. A cache mechanism is also proposed to speed up the SSJA algorithm. The experimental results show that the SSJA algorithm performs better than the STA algorithm in term of the cost and applicability, and it can significantly improve the SSJA's performance by introducing the cache mechanism.

Performance Analysis of Cache Based on Popularity and Class in Named Data Network

The communication network is growing with some unique characteristics, such as consumers repeatedly request the same content to the server, similarity in local demand trend, and dynamic changes to requests within a specific period. Therefore, a different network paradigm is needed to replace the IP network, namely Named Data Network (NDN). The content store, which acts as a crucial component in the NDN nodes is a limited resource. In addition, a cache mechanism is needed to optimize the router’s content store by exploiting the different content services characters in the network. This paper proposes a new caching algorithm called Cache Based on Popularity and Class (CAPIC) with dynamic mechanism, and the detail explanation about the static method also presented. The goal of Static-CAPIC was to enhance the total cache hit ratio on the network by pre-determining the cache proportion for each content class. However, this technique is not appropriate to control the cache hit ratio for priority class. Therefore, the Dynamic-CAPIC is used to provide flexibility to change the cache proportion based on the frequency of requests in real-time. The formula involves considering the consumers’ request all the time. It gives a higher cache hit ratio for the priority content class. This method outperforms Static-CAPIC, and the LCD+sharing scheme in the total network cache hit ratio parameter and channels it to the priority class.

CCN-Based Inter-Vehicle Communication for Efficient Collection of Road and Traffic Information

Recently, inter-vehicle communication, which helps to avoid collision accidents (by driving safety support system) and facilitate self-driving (by dissemination of road and traffic information), has attracted much attention. In this paper, in order to efficiently collect road/traffic information in the request/response manner, first a basic method, Content-centric network (CCN) for Vehicular network (CV), is proposed, which applies CCN cache function to inter-vehicle communication. Content naming and routing, which take vehicle mobility into account, are investigated. On this basis, the CV method is extended (called ECV) to avoid the cache miss problem caused by vehicle movement, and is further enhanced (called ECV+) to more efficiently exploit cache buffer in vehicles, caching content according to a probability decided by a channel usage rate. Extensive evaluations on the network simulator Scenargie, with a realistic open street map, confirm that the CV method and its extensions (ECV, ECV+) effectively reduce the average number of hops of data packets (by up to 47%, 63%, and 83%, respectively) and greatly improve the content acquisition success rate (by up to 356%, 444%, and 689%, respectively), compared to the method without a cache mechanism.

Privacy Protection of Social Networks Based on Classified Attribute Encryption

With the rapid development of social networks, privacy has also attracted attention. Based on this problem, a privacy protection scheme for social networks based on classified attribute encryption (PPSSN) is proposed for the data owner and attribute management server to manage user permissions; the approach reduces data owner overhead and also avoids use of a property management server to limit access user collusion attacks. To balance the privacy and security of data publication, this scheme classifies users and designs access control for different users and different privileges. In addition, this paper also introduces a good friend data cache mechanism to improve and optimize the original scheme to reduce the cost of decryption. The efficiency and system overhead of the proposed scheme are compared and analyzed based on experiments. The experiments show that the proposed scheme improves query efficiency, reduces system cost, and enhances privacy security.