• 한국인터넷방송통신학회논문지
• /
• 제21권5호
• /
• pp.221-226
• /
• 2021

협력 필터링 기반의 추천 시스템은 사용자들의 평가 이력을 바탕으로 하여 현 사용자가 선호할 만한 상품들을 추천해 주며 현재 다양한 상업용 목적의 필수불가결한 기능이다. 추천 상품을 결정하기 위하여, 유사한 평가 이력을 기반으로 미평가 상품들에 대한 선호 예측치를 산출하는데, 기존 연구에서 대개 두 가지 방법, 즉, 유사 사용자 기반 또는 유사 항목 기반 방법을 각기 개별적으로 활용해 왔다. 이들 방법들은 사용자들의 평가 데이터가 희소할 경우 또는 유사 사용자나 유사 항목을 구하기 어려울 경우에 산출한 예측치의 정확성이 저하되는 문제점이 있다. 본 연구에서는 이들 두가지 방법을 통합하여 평가치를 예측하는 새로운 방법을 제안한다. 제안 방법의 장점은 보다 많은 수의 유사 평가치들을 참조할 수 있으므로 추천의 질이 향상된다는 점이다. 성능 실험 결과 제안 방법은 희소한 데이터셋에서 예측치 정확도, 추천 항목 적합도, 항목 순위 적합도의 모든 측면에서 기존 방법의 성능을 크게 향상시켰으며, 다소 밀집한 데이터셋에서는 예측치 정확도 측면에서는 가장 우수하고, 다른 평가 척도에서는 기존 방법과 대등한 결과를 보였다.

• 대한한의학원전학회지
• /
• 제32권1호
• /
• pp.61-74
• /
• 2019

Objectives : The purpose of this study is to help select an appropriate word embedding method when analyzing East Asian traditional medicine texts as data. Methods : Based on prescription data that imply traditional methods in traditional East Asian medicine, we have examined 4 count-based word embedding and 2 prediction-based word embedding methods. In order to intuitively compare these word embedding methods, we proposed a "prescription generating game" and compared its results with those from the application of the 6 methods. Results : When the adjacent vectors are extracted, the count-based word embedding method derives the main herbs that are frequently used in conjunction with each other. On the other hand, in the prediction-based word embedding method, the synonyms of the herbs were derived. Conclusions : Counting based word embedding methods seems to be more effective than prediction-based word embedding methods in analyzing the use of domesticated herbs. Among count-based word embedding methods, the TF-vector method tends to exaggerate the frequency effect, and hence the TF-IDF vector or co-word vector may be a more reasonable choice. Also, the t-score vector may be recommended in search for unusual information that could not be found in frequency. On the other hand, prediction-based embedding seems to be effective when deriving the bases of similar meanings in context.

• International Journal of Aeronautical and Space Sciences
• /
• 제13권3호
• /
• pp.377-385
• /
• 2012

The Automatic Dependent Surveillance Broadcast (ADS-B) system is a key component of CNS/ATM recommended by the International Civil Aviation Organization (ICAO) as the next generation air traffic control system. ADS-B broadcasts identification, positional data, and operation information of an aircraft to other aircraft, ground vehicles and ground stations in the nearby region. This paper explores the ADS-B based trajectory prediction and the conflict detection algorithm. The multiple-model based trajectory prediction algorithm leads accurate predicted conflict probability at a future forecast time. We propose an efficient and accurate algorithm to calculate conflict probability based on approximation of the conflict zone by a set of blocks. The performance of proposed algorithms is demonstrated by a numerical simulation of two aircraft encounter scenarios.

• 대한원격탐사학회지
• /
• 제24권5호
• /
• pp.453-462
• /
• 2008

The objective of this paper is to compare the prediction performances of different landslide hazard maps based on topographic data stemming from different sources of elevation. The geostatistical framework of kriging, which can properly integrate spatial data with different accuracy, is applied for generating more reliable elevation estimates from both sparse elevation spot heights and exhaustive ASTER-based elevation values. A case study from Boeun, Korea illustrates that the integration of elevation and slope maps derived from different data yielded different prediction performances for landslide hazard mapping. The landslide hazard map constructed by using the elevation and the associated slope maps based on geostatistical integration of spot heights and ASTER-based elevation resulted in the best prediction performance. Landslide hazard mapping using elevation and slope maps derived from the interpolation of only sparse spot heights showed the worst prediction performance.

• Journal of Information Processing Systems
• /
• 제15권5호
• /
• pp.1231-1242
• /
• 2019

This paper presents a two-dimensional attention-based long short-memory (2D-ALSTM) model for stock index prediction, incorporating input attention and temporal attention mechanisms for weighting of important stocks and important time steps, respectively. The proposed model is designed to overcome the long-term dependency, stock selection, and stock volatility delay problems that negatively affect existing models. The 2D-ALSTM model is validated in a comparative experiment involving the two attention-based models multi-input LSTM (MI-LSTM) and dual-stage attention-based recurrent neural network (DARNN), with real stock data being used for training and evaluation. The model achieves superior performance compared to MI-LSTM and DARNN for stock index prediction on a KOSPI100 dataset.

• 국제학술발표논문집
• /
• The 6th International Conference on Construction Engineering and Project Management
• /
• pp.349-352
• /
• 2015

Size of building has a direct relationship with building cost, energy use and space maintenance cost. Therefore, minimizing building size during a project development is of paramount importance against such wastes. However, incautious reduction of building size may result in crowded space, and therefore harms the functionality despite the fact that building is supposed to satisfactorily support users' activity. A well-balanced design solution is, therefore, needed at an optimum level that minimizes building size in tandem with providing sufficient space to maintain functionality. For such design, architects and engineers need to be informed accurate and reliable space-use information. We present in this paper a conceptual framework of an agent-based space-use prediction simulation system that provides individual level space-use information over time in a building in consideration of project specific user information and activity schedules, space preference, ad beavioural rules. The information will accordingly assist architects and engineers to optimize space of the building as appropriate.

• 정보처리학회논문지:컴퓨터 및 통신 시스템
• /
• 제5권12호
• /
• pp.439-446
• /
• 2016

본 논문에서는 예측기반 병렬 이벤트구동 게이트수준 타이밍 시뮬레이션의 성능 개선을 위한 효율적인 예측정확도 향상 전략을 제안한다. 제안된 기법은 병렬 이벤트구동 로컬시뮬레이션들의 입력값과 출력값에 대한 예측을 이중으로 예측할 뿐만 아니라, 특별한 상황에서는 동적으로 예측할 수 있게 한다. 이중 예측은 첫번째 예측이 틀린 경우에 두번째 정적 예측 데이터로써 새로운 예측을 시도하게 되며, 동적 예측은 실제의 병렬 시뮬레이션 실행 과정 도중에 동적으로 축적되어진 지금까지의 시뮬레이션 결과를 예측 데이터로 활용하는 것이다. 제안된 두가지의 예측정확도 향상 기법은 병렬 시뮬레이션의 성능 향상의 제약 요소인 동기 오버헤드 및 통신 오버헤드를 크게 감소시킨다. 이 두가지 중요한 예측정확도 향상 방법을 통하여 6개의 디자인들에 대한 예측기반 병렬 이벤트구동 게이트수준 타이밍 시뮬레이션이 기존 통상적 방식의 상용 병렬 멀티-코어 시뮬레이션에 비하여 약 5배의 시뮬레이션 성능이 향상됨을 확인할 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권7호
• /
• pp.3150-3170
• /
• 2016

In Intelligent Transportation Systems (ITS), logistics distribution and mobile e-commerce, the real-time, accurate and reliable vehicle trajectory prediction has significant application value. Vehicle trajectory prediction can not only provide accurate location-based services, but also can monitor and predict traffic situation in advance, and then further recommend the optimal route for users. In this paper, firstly, we mine the double layers of hidden states of vehicle historical trajectories, and then determine the parameters of HMM (hidden Markov model) by historical data. Secondly, we adopt Viterbi algorithm to seek the double layers hidden states sequences corresponding to the just driven trajectory. Finally, we propose a new algorithm (DHMTP) for vehicle trajectory prediction based on the hidden Markov model of double layers hidden states, and predict the nearest neighbor unit of location information of the next k stages. The experimental results demonstrate that the prediction accuracy of the proposed algorithm is increased by 18.3% compared with TPMO algorithm and increased by 23.1% compared with Naive algorithm in aspect of predicting the next k phases' trajectories, especially when traffic flow is greater, such as this time from weekday morning to evening. Moreover, the time performance of DHMTP algorithm is also clearly improved compared with TPMO algorithm.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권9호
• /
• pp.2101-2119
• /
• 2013

The spectrum allocation is an attractive issue for mobile cognitive radio (CR) network. However, the time-varying characteristic of the spectrum allocation is not fully investigated. Thus, this paper originally deduces the probabilities of spectrum availability and interference constrain in theory under the mobile environment. Then, we propose a prediction mechanism of the time-varying available spectrum lists and the dynamic interference topologies. By considering the node mobility and primary users' (PUs') activity, the mechanism is capable of overcoming the static shortcomings of traditional model. Based on the mechanism, two prediction-based spectrum allocation algorithms, prediction greedy algorithm (PGA) and prediction fairness algorithm (PFA), are presented to enhance the spectrum utilization and improve the fairness. Moreover, new utility functions are redefined to measure the effectiveness of different schemes in the mobile CR network. Simulation results show that PGA gets more average effective spectrums than the traditional schemes, when the mean idle time of PUs is high. And PFA could achieve good system fairness performance, especially when the speeds of cognitive nodes are high.

• Geomechanics and Engineering
• /
• 제13권5호
• /
• pp.845-860
• /
• 2017

It is very important to study displacement prediction in geotechnical engineering. Nowadays, the grey system method, time series analysis method and artificial neural network method are three main methods. Based on the brief introduction, the three methods are analyzed comprehensively. Their merits and demerits, applied ranges are revealed. To solve the shortcomings of the artificial neural network method, a new prediction method based on new evolutionary neural network is proposed. Finally, through two real engineering applications, the analysis of three main methods and the new evolutionary neural network method all have been verified. The results show that, the grey system method is a kind of exponential approximation to displacement sequence, and time series analysis is linear autoregression approximation, while artificial neural network is nonlinear autoregression approximation. Thus, the grey system method can suitably analyze the sequence, which has the exponential law, the time series method can suitably analyze the random sequence and the neural network method almostly can be applied in any sequences. Moreover, the prediction results of new evolutionary neural network method is the best, and its approximation sequence and the generalization prediction sequence are all coincided with the real displacement sequence well. Thus, the new evolutionary neural network method is an acceptable method to predict the measurement displacements of geotechnical engineering.