초록 ▼

본 연구소의 선행 연구인 “기상콘텐츠 제공 및 이용을 위한 클라우드 기반 시스템 구축 연구”(이하 기상 클라우드연구)를 통해서 본 연구소에서는 클라우드관련 많은 기술들을 습득하고 보유하고 있다.
기상 클라우드 연구를 진행하면서 클라우드 시스템에서 대기모델을 구동하는 시험을 진행하였고, 이 시험을 통해서 과학전문 클라우드(아마존의 고사양 EC)를 사용할 경우에는 가능성이 있지만, 조금은 부정적인 연구 결과를 얻었다. 하지만, 대기 모델의 구동 결과를 기반으로 하는 통계 모델이나 수문/농업 등에서 진행하는 응용 기상/기후 모델의

본 연구소의 선행 연구인 “기상콘텐츠 제공 및 이용을 위한 클라우드 기반 시스템 구축 연구”(이하 기상 클라우드연구)를 통해서 본 연구소에서는 클라우드관련 많은 기술들을 습득하고 보유하고 있다.
기상 클라우드 연구를 진행하면서 클라우드 시스템에서 대기모델을 구동하는 시험을 진행하였고, 이 시험을 통해서 과학전문 클라우드(아마존의 고사양 EC)를 사용할 경우에는 가능성이 있지만, 조금은 부정적인 연구 결과를 얻었다. 하지만, 대기 모델의 구동 결과를 기반으로 하는 통계 모델이나 수문/농업 등에서 진행하는 응용 기상/기후 모델의 경우는 클라우드 기반에서도 구동이 가능하다는 것을 연구하게 되었다.
이에 따라서 그 후속 연구로 클라우드 기반 기상/기후 응용 모델 구동 및 개발 연구를 진행하고자 한다.
본 연구에서 연구소는 2가지 주제를 연구하고, 그 연구 결과를 본 연구소의 클라우드 상에서 구동하고, 판매하는 것을 목표로 한다.
그 첫 번째는 연구소의 기상 데이터 수신 서버에서 데이터를 수집하고, 연구를 진행하고자 하는 음식점과 연계하여 음식점 매출 정보 수집 시스템을 구축하고, 이를 바탕으로 Hadoop MapReduce를 이용하여 음식 산업과 기상과의 상관 관계를 연구하고자 한다.
두 번째는 역시 기상 클라우드 연구에서 등록한 특허인 “기상정보를 이용한 산업동향정보 제공장치(특허 제 10-1223070호)”에서 필요로 하는 응용 모델을 개발하고자 한다.
1. Hadoop MapReduce 심화 연구
2. 기상데이타와 음식 산업 매출 정보 매시업 연구
- 클라우드 기상 데이타 수신 시스템 사용
- 매출 정보 수신 시스템 작성
- 과거 매출 정보 분석 시스템 작성
- 과거 날씨와 과거 매출 정보를 이용한 지표 생성
3. Hadoop MapReduce를 이용한 음식산업분야 응용 모델 개발
2번 연구를 통한 지표와 미래 날씨 정보를 이용한 예측 알고리즘 개발
- Hadoop MapReduce를 이용한 음식 산업분야 응용 모델 개발

Abstract ▼

Chapter 4 The result of research and development Clause 1 What is Big data?
1. Summury
Big Data is a big data that literally. At the definition of a Wiki, Big Data means as follows :
Big Data usually includes data sets with sizes beyond the ability of commonly used software tools to capture

Chapter 4 The result of research and development Clause 1 What is Big data?
1. Summury
Big Data is a big data that literally. At the definition of a Wiki, Big Data means as follows :
Big Data usually includes data sets with sizes beyond the ability of commonly used software tools to capture, curate, manage, and process the data within a tolerable elapsed time. Big data sizes are a constantly moving target, as of 2012 ranging from a few dozen terabytes to many petabytes of data in a single data set. Big data can also be defined as "Big data is a large volume unstructured data which can not be handled by standard database management systems like DBMS, RDBMS or ORDBMS".
In 2012, Gartner updated its definition as follows: "Big data is high volume, high velocity, and/or high variety information assets that require new forms of processing to enable enhanced decision making, insight discovery and process optimization."Additionally, a new V "Veracity" is added by some organizations to describe it.
2. Implement Technology
To process a big data, it is needed a variety implemented skills. First of all, First of all, it consists of the laaS for storing large amounts of data, and consists of the Data Import & Export for importing and exporting data from outer, and consists of the calculator and process tool for dealing big data, and consists of the database for storing data, and consists of the work flow management tool for draining a data and processing a data, and consists of the decision-making used this big data, and consists of the BI (Business Intelligence) and CI (Collective Intelligence) for data analysis, and consists of the big data analysis system for supporting machine running etc, and last consists of big data management system for managing these tools.
In this study, we will find out more about the Data Import & Export, the calculator and processing tool for big data, the data storage, the work flow management tool, the big data analysis system and big data management system, and exclude the hardware part, laaS.
Clause 2 Hadoop further research
1. Summury
The big data market is growing day by day. Capacity is growing day by day, the hardware industry is more cheaper. It is possible to store a variety data that it couldn't store in past. For example, it was stored a statistic data log for statistics data only, but it wasn't stored a its variety log data. And todays, it is storing a variety log data and original data for the reanalysis.
In IT industry, The Hadoop is always mentioned when talking the big data. But, reality is Hadoop != Big Data. In 21 century, the Hadoop is a digital all-purpose tool and platform for big data. It means that the Hadoop is used by processing the big data, but the big data is not the Hadoop.
We can find the Hadoop in Wiki as follows definition :
The Hadoop is the software framework supporting a distributed application programs to operate in a big computer cluster to deal with a big data. The Hadoop is implemented based on the document of Google File System (GFS) for the Hadoop Distributed File System (HDFS) and the MapReduce.” It has begun from yahoo at first. but now in the Apache foundation it is maintaining and developing.
When it always mentioned the Hadoop, there are two way.
First one is the Hadoop Core, and the second one is the Hadoop EcoSystem. The Hadoop Core is the system based the Hadoop. and it means the HDFS and Hadoop MapReduce FrameWork.
The Hadoop EcoSystem is the Hadoop ecosystem about variety big data story. and it contains a variety systems like a Oozie, Flume, Sqoop etc.
Clause 3 Research of sale information mashup about weather data and food industry
1. Summury
We collected the weather information, sale information, social data.
and we performed the series of processes to making the standard for weather data.
We used the weekly forecasting information and observation information of the KMA. and we used the sale information of three different company. and we also used the social data that was written in the case of Twitter posts.
2. Collect the weather data
We used the weather data of KMA's observation data and weekly forecast data.
3. Collect sale data
We used the sale data of three different company. These are the one of Grilled Pork Belly restaurant, and the one of coffee shop, and the one of normal restaurant. We received the excel data files from the Grilled Pork Belly restaurant and the coffee shop. And we entered the data from the normal restaurant equipment manually.
The Grilled Pork Belly restaurant was provided the oneday sale data with excel file. The coffee shop was provided the POS data. The restaurant was provided the book of sale record.
4. Collect social data
We developed the analysis tools using the data from twitter API. So we were collecting the restaurant sale data from tweeter API. but, the collected data was not enough. so, first we were collecting the Grilled Pork Belly restaurant and the coffee shop, and we analyzed the restaurant's name into tweeter.
5. Mashup of variety data
We were analyzing the data of each restaurant analysis, social, weather information by mashup.
Clause 4 Develop the model using Hadoop MapReduce
1. Summury
We were used Hadoop MapReduce to producing the results. Program for inputting data was made by php language with MySQL database Engine. We were using Sqoop for collecting data from MySQL.
We analyzed weather information with hadoop mapreduce and hdfs. We used the Sqoop for transferring result data to MySQL.
2. Creating the standard data for model We needed two information for getting Model Base Information Data. These were the
Weather Information data and the Sales Information data. The coffee shop was sent the 3-year POS(Point Of Sales) data to us, The Grilled Pork Belly restaurant was send the 5-year daily data to us by excel and the general restaurant was send the book of 7-year sale record to us. We could get the KMA weather data since 1983 from 60 station.
3. Runnig model
We were using information to various criteria with weighting. and we were forecasted the probability forecast with Upper, Normal, Below.
4. Model validation
We were verified the model with observation data. Check points were total sales average, 3 month sales average, 1 month sales average. And we calculated these data and real daily sales data.
Clause 5 Menu suggestion system for sales promotion by weather condition
1. Summury
we were filed a patent application, 『Menu suggestion System for Sales Promotion by weather condition』 by Result From 1s Chapter to 4th Chapter. The CEO for Food Sales is inclined to a lot of effort. For example, remodeling the building, developing the menus. These sales promotion efforts are learnt by experience. The experiences are not objective. That efforts are not to be increased sales. This invention is for sales promotion about menu suggestion system by weather condition. For more detail, we analysis a food supply and demand information, Store’s food and drink sales/stock information and weather information. The technique that we forecast to suggest menu by weather condition.
2. Contetns of invention
This is menu suggestion system for sales promotion by weather condition. We analysis weather information, The social Information, the price information, the store information, the menu information and the sales information Data. And then we make and run model. We analysis model result. we forecast sales trends.
This invention to be solved is not limited by upper sentence. and those skilled in the art are understood problem to be not mentioned by below:
The Menu Suggestion System for Sales Promotion include
The part of collection for weather information.
The price information, the store information, the menu information, the sales information, etc.
The part of analysis for Information with collected data.
The part of analysis to integrate information with making based information by weather condition.
The part of model for sales predict with base information by weather condition.
The part of model for menu suggestion by weather condition.
And include
The part of verification for two models with forecasted data and real data

목차 Contents

• 표지 ... 1
• 제 출 문 ... 2
• 보고서 요약서 ... 3
• 요 약 문 ... 5
• SUMMURY ... 11
• Contents ... 20
• Tables ... 21
• Figures ... 22
• 목차 ... 24
• 표 차례 ... 25
• 그림 차례 ... 26
• 제 1 장 연구개발 과제의 개요 ... 28
• 제 1절 연구사업명 ... 30
• 제 2절 연구개발 목적 ... 30
• 제 3절 연구개발의 필요성 ... 30
• 제 4절 연구개발의 범위 ... 32
• 제 2 장 국내외 기술개발 현황 ... 34
• 제 1절 세계수준 ... 36
• 제 2절 국내수준 ... 37
• 제 3 장 연구개발 결과 ... 38
• 제 1절 Big Data란 무엇인가 ... 40
• 1. 개요 ... 40
• 제 2절 Hadoop 심화연구 ... 49
• 제 3절 기상 데이터와 음식산업 매출정보 매쉬업 연구 ... 59
• 1. 개요 ... 59
• 2. 기상 데이터 수집 ... 59
• 3. 매출 데이터 수집 ... 60
• 4. 소셜 데이터 수집 ... 62
• 5. 기상 - 매출 데이터 매쉬업 ... 62
• 제 4절 Hadoop MapReduce를 이용한 응용모델 개발 ... 70
• 1. 개요 ... 70
• 2. 모델 기준데이터 작성 ... 70
• 3. 모델 구동 ... 70
• 4. 모델 검증 ... 71
• 제 5절 매출 증진을 위한 날씨별 추천메뉴 제공 시스템 ... 73
• 1. 개요 ... 73
• 2. 발명의 내용 ... 73
• 제 6절 커피 메뉴 추천 시스템 개발 ... 74
• 1. 개요 ... 74
• 2. 시스템 구성도 ... 74
• 3. 실제 시스템 ... 74
• 제 4 장 목표달성도 및 관련분야의 기여도 ... 78
• 제 1절 연구개발 목표 달성도 ... 80
• 제 2절 관련 분야 기여도 ... 81
• 1. 기술적 측면 ... 81
• 2. 경제적·산업적 측면 ... 81
• 제 5 장 연구개발결과의 활용계획 ... 82
• 제 1절 연구개발 결과 활용 방안 ... 84
• 1. 연구개발 결과 활용 방안 ... 84
• 2. 추가연구 필요성 ... 84
• 제 6 장 연구개발과정에서 수집한 해외과학기술정보 ... 86
• 제 1절 개요 ... 88
• 제 2절 Google File Sysyem ... 88
• 제 3절 MapReduce ... 88
• 제 4절 Hadoop ... 89
• 제 5절 Hadoop Echosystem ... 90
• 제 7 장 연구시설 및 장비 현황 ... 92
• 제 1절 연구시설 현황 ... 94
• 제 2절 연구장비 현황 ... 94
• 1. 연구장비 시스템 구성도 ... 94
• 2. 연구장비 목록 ... 95
• 제 8 장 참고문헌 ... 96
• 별첨 ... 100
• 끝페이지 ... 184