• Korean Journal of Construction Engineering and Management
• /
• v.6 no.2 s.24
• /
• pp.80-91
• /
• 2005

Although construction site photos contain important as_built information, technique and knowledge, there has been lack of frameworks to store and manage construction site photos efficiently and effectively. The problems in site photo management are getting increasingly serious, as digital cameras are adapted as collection tools of site Photos. This research suggests an information framework(named CIIM: Construction Image Information Model) to manage and share construction information based on 5W1H in order to derive construction context, which includes technologies, lessons-teamed and knowledge, from construction site photos, and a site photo management system named CIMS II (Construction Image information Management system II was developed to verify the model. It is expected that the results of this research that are an information framework and an system could help more effective classification, management, search and derivation of context in a construction project.

• KIISE Transactions on Computing Practices
• /
• v.23 no.7
• /
• pp.446-451
• /
• 2017

The statistical context-sensitive spelling correction technique in this thesis is based upon Shannon's noisy channel model. The interpolation method is used for the improvement of the correction method proposed in the paper, and the general interpolation method is to fill the middle value of the probability by (N-1)-gram and (N-2)-gram. This method is based upon the same statistical corpus. In the proposed method, interpolation is performed using the frequency information between the statistical corpus and the correction document. The advantages of using frequency of correction documents are twofold. First, the probability of the coined word existing only in the correction document can be obtained. Second, even if there are two correction candidates with ambiguous probability values, the ambiguity is solved by correcting them by referring to the correction document. The method proposed in this thesis showed better precision and recall than the existing correction model.

• Journal of Advanced Navigation Technology
• /
• v.12 no.5
• /
• pp.470-482
• /
• 2008

This paper propose role base access control model that use context information for ubiquitous environment. Concept of access control that use context information assigns permission that can approach in some information or object in part. And do so that can assigned user in part to it and acquire permission. So it can approach in information or object. Therefore, user approaches in information or object in assigned role, and the role that is allocated ro own is having. So, do so that can secure information or utilization of object safety. Proposa1 model investigated lacking restriction item in GEO-RBAC model. So, it considered that present new restriction condition and role conflict in various case. Also, to GEO-RBAC model proposed suitable model, analyzed old model's advantage, shortcoming. And it presented proposal model to GEO-RBAC because improving this.

• Journal of Korea Multimedia Society
• /
• v.8 no.8
• /
• pp.1088-1098
• /
• 2005

Context-Aware is a one of the important researches in ubiquitous computing for providing optimal service to users by acquiring user's intentions and environmental information. Diverse researches are focused on the users and its environment facts for Context-Aware base and introduce a sensing based middleware which engages sever/sensor that operates identifier information to provide services. Context-Aware service which is limited by users and environment facts has the problem of overlapping sensing, unnecessary searching and anonymity of users. Also Server-Centric Context-Aware system requires very high cost to manage and operate the services. On this paper, We introduce Autogenous Context service model to simplify the Context-Aware process and design the middleware which performs decentralize management for Context-Aware information of user's portable devices to minimize problems which is occurred during the management and operation of existing Context-Aware system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.11
• /
• pp.2839-2852
• /
• 2013

Context-awareness is an essential part of ubiquitous computing, and over the past decade video based activity recognition (VAR) has emerged as an important component to identify user's context for automatic service delivery in context-aware applications. The accuracy of VAR significantly depends on the performance of the employed human body segmentation algorithm. Previous human body segmentation algorithms often engage modeling of the human body that normally requires bulky amount of training data and cannot competently handle changes over time. Recently, active contours have emerged as a successful segmentation technique in still images. In this paper, an active contour model with the integration of Chan Vese (CV) energy and Bhattacharya distance functions are adapted for automatic human body segmentation using depth cameras for VAR. The proposed technique not only outperforms existing segmentation methods in normal scenarios but it is also more robust to noise. Moreover, it is unsupervised, i.e., no prior human body model is needed. The performance of the proposed segmentation technique is compared against conventional CV Active Contour (AC) model using a depth-camera and obtained much better performance over it.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.6
• /
• pp.745-751
• /
• 2004

In conventional boolean retrieval systems, document ranking is not supported and similarity coefficients cannot be computed between queries and documents. The MMM(Max and Min Model), Paice and P-norm models have been proposed in the past to support the ranking facility for boolean retrieval systems. They have common properties of interpreting boolean operators softly In this paper we propose a new soft evaluation method for web Information retrieval using local context analysis feedback model. We also show through performance comparison that local contort analysis feedback is more efficient and effective than MMM, Paice and P-norm.

• Proceedings of the KSRS Conference
• /
• 2007.10a
• /
• pp.568-572
• /
• 2007

With the proliferation of advanced wireless network and sensor technologies, smart devices under USN(ubiquitous sensor network) environment are capable of collecting context information such as temperature, humidity, weight, and location about objects at real time. Therefore, applications must be able to analyze collected information and notify useful information to wanted users timely. This service can be realized by implementing an event-based middleware. In the middleware, event messages collected from physical environment will be filtered according to profiles that users define in advance and the result will be sent to the interested users. In this paper, we present XML-based event model, ECA-based profile model, and the architecture of an event-based middleware suitable to USN environment. We will also model and describe them using the examples of logistics area. By implementing the system based on the design above, the middleware enable applications or users to easily access to physical sources. The proposed middleware can also apply to not only logistics area but also other various areas under USN environment such as intelligent traffic control system, national disaster management system and u-medical system.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.3 s.35
• /
• pp.119-132
• /
• 2005

The health care system revolutionized by the use of information and communication technologies. Computer information processing and electronic communication technologies play an increasingly important role in the area of health care. We propose a new role based access control model for pervasive health care systems, which changed location, time, environment information. Also our model can be solved the occurrence of an reduction authority problem to pervasive health care system at emergency environment. We propose a new role based access control model for pervasive health care systems, which combines role-to-role delegations, negative permission, context concept and dynamic context aware access control. With out approach we aim to preserver the advantages of RBAC and offer groat flexibility and fine-grained access control in pervasive healthcare information systems.

• Journal of the Korean Society of Safety
• /
• v.23 no.6
• /
• pp.51-56
• /
• 2008

As RFID/USN technology is used in the latest industry trend, the information analysis paradigm shifts to intelligence service environment. The intelligent service includes autonomic operation, which select activity by defining itself to the status of industry facilities. Furthermore, information analysis based on IT used to frequently data mining for detecting the meaning information and deriving new pattern. This paper suggest self-classifying of context-aware by applying data mining in gas facilities for serving the intelligent gas safety management. We modify data algorithm for fitting the domain of gas safety, construct context-aware model by using the proposed algorithm, and demonstrate our method. As the accuracy of our model is improved over 90%, the our approach can apply to intelligent gas safety management based on RFID/USN environments.

• ETRI Journal
• /
• v.41 no.3
• /
• pp.371-382
• /
• 2019

Machine reading comprehension is the task of understanding a given context and finding the correct response in that context. A simple recurrent unit (SRU) is a model that solves the vanishing gradient problem in a recurrent neural network (RNN) using a neural gate, such as a gated recurrent unit (GRU) and long short-term memory (LSTM); moreover, it removes the previous hidden state from the input gate to improve the speed compared to GRU and LSTM. A self-matching network, used in R-Net, can have a similar effect to coreference resolution because the self-matching network can obtain context information of a similar meaning by calculating the attention weight for its own RNN sequence. In this paper, we construct a dataset for Korean machine reading comprehension and propose an $S^2-Net$ model that adds a self-matching layer to an encoder RNN using multilayer SRU. The experimental results show that the proposed $S^2-Net$ model has performance of single 68.82% EM and 81.25% F1, and ensemble 70.81% EM, 82.48% F1 in the Korean machine reading comprehension test dataset, and has single 71.30% EM and 80.37% F1 and ensemble 73.29% EM and 81.54% F1 performance in the SQuAD dev dataset.