• Journal of Information Science Theory and Practice
• /
• 제2권4호
• /
• pp.31-48
• /
• 2014

The proliferation of many interactive Topic Detection and Tracking (iTDT) systems has motivated researchers to design systems that can track and detect news better. iTDT focuses on user interaction, user evaluation, and user interfaces. Recently, increasing effort has been devoted to user interfaces to improve TDT systems by investigating not just the user interaction aspect but also user and task oriented evaluation. This study investigates the combination of the bag of words and named entities approaches implemented in the iTDT interface, called Interactive Event Tracking (iEvent), including what TDT tasks these approaches facilitate. iEvent is composed of three components, which are Cluster View (CV), Document View (DV), and Term View (TV). User experiments have been carried out amongst journalists to compare three settings of iEvent: Setup 1 and Setup 2 (baseline setups), and Setup 3 (experimental setup). Setup 1 used bag of words and Setup 2 used named entities, while Setup 3 used a combination of bag of words and named entities. Journalists were asked to perform TDT tasks: Tracking and Detection. Findings revealed that the combination of bag of words and named entities approaches generally facilitated the journalists to perform well in the TDT tasks. This study has confirmed that the combination approach in iTDT is useful and enhanced the effectiveness of users' performance in performing the TDT tasks. It gives suggestions on the features with their approaches which facilitated the journalists in performing the TDT tasks.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권7호
• /
• pp.2952-2971
• /
• 2016

Outsourcing jobs to a public cloud is a cost-effective way to address the problem of satisfying the peak resource demand when the local cloud has insufficient resources. In this paper, we studied the management of deadline-constrained bag-of-tasks jobs on hybrid clouds. We presented a binary nonlinear programming (BNP) problem to model the hybrid cloud management which minimizes rent cost from the public cloud while completes the jobs within their respective deadlines. To solve this BNP problem in polynomial time, we proposed a heuristic algorithm. The main idea is assigning the task closest to its deadline to current core until the core cannot finish any task within its deadline. When there is no available core, the algorithm adds an available physical machine (PM) with most capacity or rents a new virtual machine (VM) with highest cost-performance ratio. As there may be a workload imbalance between/among cores on a PM/VM after task assigning, we propose a task reassigning algorithm to balance them. Extensive experimental results show that our heuristic algorithm saves 16.2%-76% rent cost and improves 47.3%-182.8% resource utilizations satisfying deadline constraints, compared with first fit decreasing algorithm, and that our task reassigning algorithm improves the makespan of tasks up to 47.6%.

• 한국음향학회지
• /
• 제33권4호
• /
• pp.248-254
• /
• 2014

음향 상황 인지(acoustic context awareness)는 다양하게 발생되는 음원들로부터 어떠한 장소인지 또는 어떠한 사건이 발생하는지를 판단하는 기술로 음향 이벤트 검출 또는 인식 보다 한 단계 더 복잡한 문제이다. 기존의 상황인지 기술은 음향 이벤트 검출 또는 인식 기술에 기반하여 현재 상황을 인지하는 방법을 사용하고 있다. 하지만 이와 같은 접근 방법은 여러 음원이 동시에 발생하거나 유사한 음원이 발생하는 실제 환경에서 정확한 상황 판단이 어렵다. 특히 버스와 지하철은 승객들에 의한 잡음으로 상황을 인지하기 힘들다. 이러한 문제를 극복하기 위해 본 논문에서는 유사한 음향 이벤트가 발생하는 버스와 지하철 상황을 인식할 수 있는 Bag of Words 기반의 상황 인지 알고리즘을 연구하고 코드북 생성을 위한 특징벡터를 제안한다. 제안하는 특징벡터의 효용성은 Support Vector Machine을 이용한 실험을 통해 검증했다.

• 한국콘텐츠학회:학술대회논문집
• /
• 한국콘텐츠학회 2018년도 춘계 종합학술대회 논문집
• /
• pp.7-8
• /
• 2018

본 연구에서는 Word2vec을 머신러닝 분류기를 이용해 효율적인 오피니언 마이닝 방법을 제안한다. 본 연구의 목적을 위해 BOW(Bag-of-Words) 방법과 Word2vec방법을 이용해 속성 셋을 구성했다. 구성된 속성 셋은 Decision tree, Logistic regression, Support vector machine, Random forest를 이용해 오피니언 마이닝을 수행했다. 연구 결과, Word2vec 방법과 RF분류기가 가장 높은 정확도를 나타냈다. 그리고 Word2vec 방법이 BOW방법 보다 각 분류기에서 높은 성능을 나타냈다.

• 산업공학
• /
• 제17권1호
• /
• pp.46-55
• /
• 2004

Various characteristics of the object being lifted are known to affect the biomechanical, physiological, and psychophysical stresses. The object characteristics to be considered in the design process of lifting tasks are weight, shape, stiffness, and availability of handles and similar coupling devices. In this study, a prototype Polypropylene laminated bag with carrying handles was designed to decrease the physical stress of people who handle these bags. Physiological and psychophysical approaches as well as subjective ratings were applied to evaluate the effects of handles provided on the designed PP laminated bag. Statistical analysis showed that the VO2, heart rate, blood pressure, and Borg-RPE score for PP laminated fertilizer bag with carrying handles were significantly lower than those bags without handles. Moreover, Maximum Acceptable Lifting Endurance Time(MALET) measure, newly developed in this study, for bags with handles was significantly higher than those for bags without handles. It is thus recommended that the various types of bags and boxes be equipped with handles to reduce the musculoskeletal, physiological, psychophysical, and subjective perceived stresses.

• 한국컴퓨터정보학회논문지
• /
• 제24권2호
• /
• pp.171-177
• /
• 2019

This paper aims to find the most effective feature selection method for the sake of opinion mining tasks. Basically, opinion mining tasks belong to sentiment analysis, which is to categorize opinions of the online texts into positive and negative from a text mining point of view. By using the five product groups dataset such as apparel, books, DVDs, electronics, and kitchen, TF-IDF and Bag-of-Words(BOW) fare calculated to form the product review feature sets. Next, we applied the feature selection methods to see which method reveals most robust results. The results show that the stacking classifier based on those features out of applying Information Gain feature selection method yields best result.

• 패션비즈니스
• /
• 제24권1호
• /
• pp.15-28
• /
• 2020

The purpose of this study was to develop and propose creative smart bags in emotional e-textiles using LEDs that inform smartphone users of motion-induced luminescence and ringing of cell phones. The LED light-emitting operation tasks produced in the study were applied to each of the three design smart bags, setting the five cases of luminance by a call initiated, absent phone, rejecting answering phone, texting, and motion-induced luminescence. In the male laptop bags of LED luminous images using wappen, 10 LEDs could be separated by a total of three pins to display the luminous mode, and all 10 LEDs became a total of five luminous patterns, including all that illuminate and those that illuminate randomly. E-wappen rendered the motif a strong sense of visibility and performed six roles on phone rings and texting. To develop a women's tote bag, we did a laser cut and attached the leather strips and placed 10 triangular LEDs to form a geometric LED e-textile. It provides the possibility of transforming simple design from traditional fashion into a more interesting and various smart designs. An entertainment smart bag using graphic design was constructed by applying a tilt sensor to look like a light in the night sky by shaking and moving the bag. The graphic design and composition of LEDs indicate that LEDs and fashion item are applied in harmony rather than heterogeneous, enabling them to be applied as fashion-oriented wearable smart products.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제10권9호
• /
• pp.359-366
• /
• 2021

대한민국 건설사들은 아파트 하자 정보를 축적하고 보수작업을 관리하기 위한 시스템을 운영하는데 상당한 인력과 비용을 투자하고 있다. 본 연구에서는 하자 접수 상세내용 텍스트 데이터를 이용하여 하자 보수 시설공사에 따른 세부공종을 분류하는 머신러닝 모델을 제안한다. 두 가지 단어 임베딩(Bag-of-words, Term Frequency-Inverse Document Frequency (TF-IDF))과 두 가지 분류기(Support Vector Machine, Random Forest)를 통해 한국어로 작성된 65만건 이상의 하자 접수데이터로부터 하자보수 시설공사 세부공종을 분류했다. 특히, 이번 연구에서는 특정 시설공사(마감공사)의 9개 세부공종(가전제품, 도배공사, 도장공사, 미장공사, 석공사, 수장공사, 옥내가구공사, 주방기구공사, 타일공사)을 분류하는 이진분류 모델과 다중 분류 모델을 연구했다. 그 결과, TF-IDF와 Random Forest를 사용한 두가지 분류 모델에서 90%이상의 정확도, 정밀도, 재현율 및 F1점수를 확인했다.

• 한국식품저장유통학회지
• /
• 제24권7호
• /
• pp.915-922
• /
• 2017

양파의 관행 수확과 저장 방식을 개선하여 벌크 상태로 수확하여 수송, 하역, 보관 및 저장할 수 있도록 톤백 및 와이어철제파렛트를 개발한 후 관행저장 방식과 벌크저장 후 저장성을 비교하고 그에 따른 노동투입시간 및 소요비용을 비교하여 양파의 산지물류비 절감 효과를 비교 분석하고자 하였다. 본 연구를 통해 개발된 톤백이 관행 톤백보다 길이방향 인장강도가 16% 높았으며, 개발한 와이어철제파렛트가 관행철제파렛트보다 약 10% 더 저장가능하고 차량 적재 효율도 2배 이상 향상되었다. 개발된 500 kg용 와이어 철제파렛트의 벌크저장과 관행 그물망 저장의 감모율에는 차이가 없었으나, 1,000 kg 이상 많은 양을 저장할 경우 와이어철제파렛트 벌크저장은 관행저장보다 감모율이 3.7% 높아 적합하지 않는 것으로 나타났다. 톤백을 이용한 벌크 수확이 관행 망 수확보다 노동시간이 50.1% 그리고 총 투입비용은 46.1% 감소할 수 있는 것으로 나타났다. 와이어철제파렛트를 적용한 벌크저장이 관행저장보다 총 저장비용이 28.8% 감소하였으며, 2016년 양파생산량(1,298,749톤)의 30%를 와이어철제파렛트 벌크저장으로 대체할 경우 연간 183억 원을 절감할 것으로 추정된다. 따라서 현재 산지업체들의 와이어철제파렛트를 조기 도입하여 정착시키면 전국적으로 노동력 및 비용 절감에 상당한 효과가 있을 것으로 판단된다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제16권2호
• /
• pp.104-110
• /
• 2016

Dense local image descriptors like SIFT are fruitful for capturing salient information about image, shown to be successful in various image-related tasks when formed in bag-of-words representation (i.e., histograms). In this paper we consider to utilize these dense local descriptors in the object tracking problem. A notable aspect of our tracker is that instead of adopting a point estimate for the target model, we account for uncertainty in data noise and model incompleteness by maintaining a distribution over plausible candidate models within the Bayesian framework. The target model is also updated adaptively by the principled Bayesian posterior inference, which admits a closed form within our Dirichlet prior modeling. With empirical evaluations on some video datasets, the proposed method is shown to yield more accurate tracking than baseline histogram-based trackers with the same types of features, often being superior to the appearance-based (visual) trackers.