• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.161-177
• /
• 2019

In this paper, we study the performance improvement of the answer extraction in Question-Answering system by using sentence dependency parsing result. The Question-Answering (QA) system consists of query analysis, which is a method of analyzing the user's query, and answer extraction, which is a method to extract appropriate answers in the document. And various studies have been conducted on two methods. In order to improve the performance of answer extraction, it is necessary to accurately reflect the grammatical information of sentences. In Korean, because word order structure is free and omission of sentence components is frequent, dependency parsing is a good way to analyze Korean syntax. Therefore, in this study, we improved the performance of the answer extraction by adding the features generated by dependency parsing analysis to the inputs of the answer extraction model (Bidirectional LSTM-CRF). The process of generating the dependency graph embedding consists of the steps of generating the dependency graph from the dependency parsing result and learning the embedding of the graph. In this study, we compared the performance of the answer extraction model when inputting basic word features generated without the dependency parsing and the performance of the model when inputting the addition of the Eojeol tag feature and dependency graph embedding feature. Since dependency parsing is performed on a basic unit of an Eojeol, which is a component of sentences separated by a space, the tag information of the Eojeol can be obtained as a result of the dependency parsing. The Eojeol tag feature means the tag information of the Eojeol. The process of generating the dependency graph embedding consists of the steps of generating the dependency graph from the dependency parsing result and learning the embedding of the graph. From the dependency parsing result, a graph is generated from the Eojeol to the node, the dependency between the Eojeol to the edge, and the Eojeol tag to the node label. In this process, an undirected graph is generated or a directed graph is generated according to whether or not the dependency relation direction is considered. To obtain the embedding of the graph, we used Graph2Vec, which is a method of finding the embedding of the graph by the subgraphs constituting a graph. We can specify the maximum path length between nodes in the process of finding subgraphs of a graph. If the maximum path length between nodes is 1, graph embedding is generated only by direct dependency between Eojeol, and graph embedding is generated including indirect dependencies as the maximum path length between nodes becomes larger. In the experiment, the maximum path length between nodes is adjusted differently from 1 to 3 depending on whether direction of dependency is considered or not, and the performance of answer extraction is measured. Experimental results show that both Eojeol tag feature and dependency graph embedding feature improve the performance of answer extraction. In particular, considering the direction of the dependency relation and extracting the dependency graph generated with the maximum path length of 1 in the subgraph extraction process in Graph2Vec as the input of the model, the highest answer extraction performance was shown. As a result of these experiments, we concluded that it is better to take into account the direction of dependence and to consider only the direct connection rather than the indirect dependence between the words. The significance of this study is as follows. First, we improved the performance of answer extraction by adding features using dependency parsing results, taking into account the characteristics of Korean, which is free of word order structure and omission of sentence components. Second, we generated feature of dependency parsing result by learning - based graph embedding method without defining the pattern of dependency between Eojeol. Future research directions are as follows. In this study, the features generated as a result of the dependency parsing are applied only to the answer extraction model in order to grasp the meaning. However, in the future, if the performance is confirmed by applying the features to various natural language processing models such as sentiment analysis or name entity recognition, the validity of the features can be verified more accurately.

• Journal of Intelligence and Information Systems
• /
• v.26 no.4
• /
• pp.87-109
• /
• 2020

Currently, thanks to the major stride made in developing wired and wireless communication technology, a variety of IT services are available on land. This trend is leading to an increasing demand for IT services to vessels on the water as well. And it is expected that the request for various IT services such as two-way digital data transmission, Web, APP, etc. is on the rise to the extent that they are available on land. However, while a high-speed information communication network is easily accessible on land because it is based upon a fixed infrastructure like an AP and a base station, it is not the case on the water. As a result, a radio communication network-based voice communication service is usually used at sea. To solve this problem, an additional frequency for digital data exchange was allocated, and a ship ad-hoc network (SANET) was proposed that can be utilized by using this frequency. Instead of satellite communication that costs a lot in installation and usage, SANET was developed to provide various IT services to ships based on IP in the sea. Connectivity between land base stations and ships is important in the SANET. To have this connection, a ship must be a member of the network with its IP address assigned. This paper proposes a SANET-CC protocol that allows ships to be assigned their own IP address. SANET-CC propagates several non-overlapping IP addresses through the entire network from land base stations to ships in the form of the tree. Ships allocate their own IP addresses through the exchange of simple requests and response messages with land base stations or M-ships that can allocate IP addresses. Therefore, SANET-CC can eliminate the IP collision prevention (Duplicate Address Detection) process and the process of network separation or integration caused by the movement of the ship. Various simulations were performed to verify the applicability of this protocol to SANET. The outcome of such simulations shows us the following. First, using SANET-CC, about 91% of the ships in the network were able to receive IP addresses under any circumstances. It is 6% higher than the existing studies. And it suggests that if variables are adjusted to each port's environment, it may show further improved results. Second, this work shows us that it takes all vessels an average of 10 seconds to receive IP addresses regardless of conditions. It represents a 50% decrease in time compared to the average of 20 seconds in the previous study. Also Besides, taking it into account that when existing studies were on 50 to 200 vessels, this study on 100 to 400 vessels, the efficiency can be much higher. Third, existing studies have not been able to derive optimal values according to variables. This is because it does not have a consistent pattern depending on the variable. This means that optimal variables values cannot be set for each port under diverse environments. This paper, however, shows us that the result values from the variables exhibit a consistent pattern. This is significant in that it can be applied to each port by adjusting the variable values. It was also confirmed that regardless of the number of ships, the IP allocation ratio was the most efficient at about 96 percent if the waiting time after the IP request was 75ms, and that the tree structure could maintain a stable network configuration when the number of IPs was over 30000. Fourth, this study can be used to design a network for supporting intelligent maritime control systems and services offshore, instead of satellite communication. And if LTE-M is set up, it is possible to use it for various intelligent services.