ΊέΊέί£shows by User: ssuserf03d2b / http://www.slideshare.net/images/logo.gif ΊέΊέί£shows by User: ssuserf03d2b / Mon, 22 Oct 2018 04:26:45 GMT ΊέΊέί£Share feed for ΊέΊέί£shows by User: ssuserf03d2b LDA : latent Dirichlet Allocation (Fairies NLP Series) - Korean Ver. /slideshow/lda-latent-dirichlet-allocation-fairies-nlp-series-korean-ver/120255090 fairieslda-181022042645
*Introduction - Unsupervised Learning (Text-mining or Machine learning? *Method - Learning Process, Packages *Explanation Formula *Case Study of LDA *Coding with Best LDA Model from Grid search *Conclusion - insight & Furthur more *Not edit here -What is Text-Rank? -What is Jieba Packages?]]>
*Introduction - Unsupervised Learning (Text-mining or Machine learning? *Method - Learning Process, Packages *Explanation Formula *Case Study of LDA *Coding with Best LDA Model from Grid search *Conclusion - insight & Furthur more *Not edit here -What is Text-Rank? -What is Jieba Packages?]]> Mon, 22 Oct 2018 04:26:45 GMT /slideshow/lda-latent-dirichlet-allocation-fairies-nlp-series-korean-ver/120255090 ssuserf03d2b@slideshare.net(ssuserf03d2b) LDA : latent Dirichlet Allocation (Fairies NLP Series) - Korean Ver. ssuserf03d2b *Introduction - Unsupervised Learning (Text-mining or Machine learning? *Method - Learning Process, Packages *Explanation Formula *Case Study of LDA *Coding with Best LDA Model from Grid search *Conclusion - insight & Furthur more *Not edit here -What is Text-Rank? -What is Jieba Packages? <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/fairieslda-181022042645-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> *Introduction - Unsupervised Learning (Text-mining or Machine learning? *Method - Learning Process, Packages *Explanation Formula *Case Study of LDA *Coding with Best LDA Model from Grid search *Conclusion - insight &amp; Furthur more *Not edit here -What is Text-Rank? -What is Jieba Packages?
LDA : latent Dirichlet Allocation (Fairies NLP Series) - Korean Ver. from Adonis Han
]]> 1964 3 https://cdn.slidesharecdn.com/ss_thumbnails/fairieslda-181022042645-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 (Kor ver.)NLP embedding(word2vec) tutorial & implementation(Tensorflow) /slideshow/kor-vernlp-embeddingword2vec-tutorial-implementationtensorflow/90102946 nlptutorialfairies-180309032040
NLP embedding(word2vec) tutorial & implementation # word embedding # skip-gram # Negative Sampling implementation -tensorflow -python -https://github.com/AdonisHan/nlp_word2vec_tensorflow ]]>
NLP embedding(word2vec) tutorial & implementation # word embedding # skip-gram # Negative Sampling implementation -tensorflow -python -https://github.com/AdonisHan/nlp_word2vec_tensorflow ]]> Fri, 09 Mar 2018 03:20:40 GMT /slideshow/kor-vernlp-embeddingword2vec-tutorial-implementationtensorflow/90102946 ssuserf03d2b@slideshare.net(ssuserf03d2b) (Kor ver.)NLP embedding(word2vec) tutorial & implementation(Tensorflow) ssuserf03d2b NLP embedding(word2vec) tutorial & implementation # word embedding # skip-gram # Negative Sampling implementation -tensorflow -python -https://github.com/AdonisHan/nlp_word2vec_tensorflow <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/nlptutorialfairies-180309032040-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> NLP embedding(word2vec) tutorial &amp; implementation # word embedding # skip-gram # Negative Sampling implementation -tensorflow -python -https://github.com/AdonisHan/nlp_word2vec_tensorflow
(Kor ver.)NLP embedding(word2vec) tutorial & implementation(Tensorflow) from Adonis Han
]]> 566 23 https://cdn.slidesharecdn.com/ss_thumbnails/nlptutorialfairies-180309032040-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 how to understand and implement the "WAVENET" /ssuserf03d2b/how-to-understand-and-implement-the-wavenet fairies-wavenet-180309031321
how to understand and implement the "WAVENET" Introduction -WaveNet: deep generative model of audio data that operate directly at the waveform level Contributions Method Causal convolutions Dilated causal convolutions Softmax distribution implementation -keras]]>
how to understand and implement the "WAVENET" Introduction -WaveNet: deep generative model of audio data that operate directly at the waveform level Contributions Method Causal convolutions Dilated causal convolutions Softmax distribution implementation -keras]]> Fri, 09 Mar 2018 03:13:21 GMT /ssuserf03d2b/how-to-understand-and-implement-the-wavenet ssuserf03d2b@slideshare.net(ssuserf03d2b) how to understand and implement the "WAVENET" ssuserf03d2b how to understand and implement the "WAVENET" Introduction -WaveNet: deep generative model of audio data that operate directly at the waveform level Contributions Method Causal convolutions Dilated causal convolutions Softmax distribution implementation -keras <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/fairies-wavenet-180309031321-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> how to understand and implement the &quot;WAVENET&quot; Introduction -WaveNet: deep generative model of audio data that operate directly at the waveform level Contributions Method Causal convolutions Dilated causal convolutions Softmax distribution implementation -keras
how to understand and implement the "WAVENET" from Adonis Han
]]> 1464 5 https://cdn.slidesharecdn.com/ss_thumbnails/fairies-wavenet-180309031321-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 [kor ver.]νŒ¨ν„΄μΈμ‹μ„ μœ„ν•œ 인곡신경망 Caps-net κ΅¬ν˜„ /slideshow/kor-ver-capsnet/87756953 capsulefinal-180210212655
νŒ¨ν„΄μΈμ‹μ„ μœ„ν•œ 인곡신경망 Caps-net κ΅¬ν˜„ capsnet in keras find the code here - https://github.com/AdonisHan/capsnet]]>
νŒ¨ν„΄μΈμ‹μ„ μœ„ν•œ 인곡신경망 Caps-net κ΅¬ν˜„ capsnet in keras find the code here - https://github.com/AdonisHan/capsnet]]> Sat, 10 Feb 2018 21:26:55 GMT /slideshow/kor-ver-capsnet/87756953 ssuserf03d2b@slideshare.net(ssuserf03d2b) [kor ver.]νŒ¨ν„΄μΈμ‹μ„ μœ„ν•œ 인곡신경망 Caps-net κ΅¬ν˜„ ssuserf03d2b νŒ¨ν„΄μΈμ‹μ„ μœ„ν•œ 인곡신경망 Caps-net κ΅¬ν˜„ capsnet in keras find the code here - https://github.com/AdonisHan/capsnet <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/capsulefinal-180210212655-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> νŒ¨ν„΄μΈμ‹μ„ μœ„ν•œ 인곡신경망 Caps-net κ΅¬ν˜„ capsnet in keras find the code here - https://github.com/AdonisHan/capsnet
[kor ver.]νŒ¨ν„΄μΈμ‹μ„ μœ„ν•œ 인곡신경망 Caps-net κ΅¬ν˜„ from Adonis Han
]]> 1010 1 https://cdn.slidesharecdn.com/ss_thumbnails/capsulefinal-180210212655-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 [kor ver.]Global GO (Bigdata-Cloud computing project - mainly in MVC model2) /ssuserf03d2b/kor-verglobal-go-bigdatacloud-computing-project-mainly-in-mvc-model2 random-171227155714
kor ver. [Abstract] κ³ κ°λ§žμΆ€ν˜• μ—¬ν–‰μ‚¬μ΄νŠΈοΌˆFront-Back-End / Bigdata μ—­λŸ‰οΌ‰ 아킀텍쳐 : Linux（CentOsοΌ‰ -＞ Java -＞ JSP -＞ Mysql-＞ Hadoop -＞ R -＞ Output λͺ©μ  - μ—¬ν–‰μƒν’ˆμ„ κ΄€λ¦¬μžμ—κ²Œ μ‚¬μš©μžκ°€ μ›ν•˜λŠ” 정보λ₯Ό μž…λ ₯ν•˜κ³  μ‹ μ²­ν•˜μ—¬ κ·Έ μ •λ³΄οΌˆλ°μ΄ν„°οΌ‰λ₯Ό κ΄€λ¦¬μžκ°€ 그에 λ§žλŠ” 여행사 νŒ¨ν‚€μ§€μƒν’ˆμ„ 제곡 ν•˜λ„λ‘ ν™ˆνŽ˜μ΄μ§€λ₯Ό λ§Œλ“€κ³  λ‘œκ·Έλ°μ΄ν„°λ‘œ μ†ŒλΉ„μžμ„±ν–₯을 λΆ„μ„ν•˜λŠ” ν”„λ‘œμ νŠΈ μ˜€μŠ΅λ‹ˆλ‹€. 1. 고객 λ§žμΆ€ν˜• μ—¬ν–‰μΆ”μ²œ μ‚¬μ΄νŠΈ μ œμž‘ 2. λ°˜μ‘ν˜• μ›Ήμ‚¬μ΄νŠΈ κ΅¬ν˜„ 3. λ°μ΄ν„°λ² μ΄μŠ€ ν™œμš© 및 λΉ„μ •ν˜• 데이터(μ•‘μ„ΈμŠ€ 둜그) 뢄석 4. λΆ„μ‚°νŒŒμΌμ‹œμŠ€ν…œ Hadoop ν™œμš© 5. μ‹€μ‹œκ°„ 톡계 및 μΈμ‚¬μ΄νŠΈ λ„μΆœ [Method & Process] λ³Έ ν”„λ‘œμ νŠΈλŠ” μ‚¬μš©μž μ€‘μ‹¬μ˜ tour ν™ˆνŽ˜μ΄μ§€λ₯Ό λ§Œλ“€μ–΄ μ—¬λŸ¬ μ†Œν”„νŠΈμ›¨μ–΄ 툴과 κΈ°λ°˜μ„ μ‘μš©ν•˜μ—¬ 섀계뢀터 μ‹€ν–‰λ‹¨κ³„κΉŒμ§€μ˜ 과정을 거쳐 ν”„λ‘œμ νŠΈλ₯Ό μ§„ν–‰ν•˜μ˜€μŠ΅λ‹ˆλ‹€. Customer-Centered Tour λΌλŠ” 주제λ₯Ό μ •ν•˜μ—¬ μ—¬ν–‰μƒν’ˆμ„ κ΄€λ¦¬μžμ—κ²Œ μ‚¬μš©μžκ°€ μ›ν•˜λŠ” 정보λ₯Ό μž…λ ₯ν•˜κ³  μ‹ μ²­ν•˜μ—¬ κ·Έ 정보(데이터)λ₯Ό κ΄€λ¦¬μžκ°€ 그에 λ§žλŠ” 여행사 νŒ¨ν‚€μ§€μƒν’ˆμ„ 제곡 ν•˜λ„λ‘ μ„€κ³„ν•˜μ˜€μŠ΅λ‹ˆλ‹€. MVCλͺ¨λΈμ„ 기점으둜 ν™˜κ²½μ„€μ • ν”„λ‘œκ·Έλž¨, μ†Œν”„νŠΈμ›¨μ–΄, 뢄석틀을 μ΄μš©ν•˜μ—¬ ν”„λ‘œμ νŠΈλ₯Ό 연ꡬ 및 μ§„ν–‰ν•˜μ˜€μœΌλ©°, Customer-Based Tour ν™ˆνŽ˜μ΄μ§€λ₯Ό μ„€κ³„ν•˜μ—¬ μ†Œν”„νŠΈμ›¨μ–΄ μ€‘μ‹¬μœΌλ‘œ JAVA와 같은 객체지ν–₯언어와 κ΄€κ³„ν˜• λ°μ΄ν„°λ² μ΄μŠ€(Oracle11g)의 ν™˜κ²½μ„€μ •μ΄λ‚˜ (Linux) μ†Œν”„νŠΈμ›¨μ–΄λ₯Ό μ—°λ™ν•˜μ—¬ λΆ„μ„ν‹€λ‘œ 데이터λ₯Ό λΆ„μ„ν•˜λŠ” ν”„λ‘œκ·Έλž˜λ°(R, ν•˜λ‘‘)을 μ΄μš©ν•˜μ—¬ μž…μΆœλ ₯ 된 데이터λ₯Ό λΆ„μ„ν•˜μ—¬ ν”„λ‘œμ νŠΈλ₯Ό μ™„λ£Œν•˜μ˜€μŠ΅λ‹ˆ. 각 ν”„λ‘œκ·Έλž¨κ³Ό μ• ν”Œλ¦¬μΌ€μ΄μ…˜νˆ΄μ„ μ†Œκ°œν•˜κ³  μ‘μš© 및 μ—°κ΅¬μ§„ν–‰μ ˆμ°¨λ₯Ό μž‘μ„±ν•˜μ˜€μŠ΅λ‹ˆλ‹€. Customer Mode μ‚¬μš©μžκ°€ 메인 νŽ˜μ΄μ§€(index)λ₯Ό 기점으둜 νšŒμ›κ°€μž… νŽ˜μ΄μ§€λ₯Ό 톡해 아이디λ₯Ό λ§Œλ“€μ–΄ νšŒμ›μΈμ¦μ„ ν•˜μ—¬ Customer mode둜 μ ‘μ†ν•©λ‹ˆλ‹€. Customer Modeμ—μ„œλŠ” ν™ˆνŽ˜μ΄μ§€ μƒμœ„μ— 더 λ§Žμ€ μΉ΄ν…Œκ³ λ¦¬λ₯Ό 찾을 수 μžˆμŠ΅λ‹ˆλ‹€. λ˜ν•œ 메인 νŽ˜μ΄μ§€ ν•˜λ‹¨μ—μ„œ μ‚¬μš©μžκ°€ μ›ν•˜λŠ” 각 λ‚˜λΌλ³„ 여행지λ₯Ό ν΄λ¦­ν•˜μ—¬ λ‹€μŒνŽ˜μ΄μ§€μΈ λ‚˜λΌλ³„μƒμ„Έμ •λ³΄λ₯Ό λ³Ό 수 있음과 λ™μ‹œμ— νŒ¨ν‚€μ§€μƒν’ˆμ„ μ‹ μ²­ν•  수 μžˆλŠ” νŽ˜μ΄μ§€λ‘œ 가도둝 링크λ₯Ό μ„€κ³„ν•˜μ˜€μŠ΅λ‹ˆλ‹€. λ˜ν•œ μ‚¬μš©μžλŠ” ν™ˆνŽ˜μ΄μ§€μ—μ„œ μ œκ³΅ν•˜λŠ” 여행지 톡계λ₯Ό 각 λ‚˜λΌλ³„, ν…Œλ§ˆλ³„, 계쑀별, μ—°λ Ήλ³„λ‘œ λ³Ό 수 μžˆλ„λ‘ μ œκ³΅λ©λ‹ˆλ‹€. Administrator Mode νŒ¨ν‚€μ§€μƒν’ˆμ„ λ“±λ‘ν•˜λ©΄ κ·Έ 정보가 κ΄€λ¦¬μž(Administrator)μ—κ²Œ λ‘œκ·Έλ°μ΄ν„°κ°€ μž…λ ₯되고 여행사별 νŒ¨ν‚€μ§€μƒν’ˆκ³Ό μ‚¬μš©μž(Customer Mode)μ—κ²Œ 연결을 ν•˜μ—¬ λ‹€μ‹œ μ‚¬μš©μžμ—κ²Œ λ³΄λ‚΄μ€λ‹ˆλ‹€. κ΄€λ¦¬μžλŠ” VIP고객정보, νšŒμ›κ΄€λ¦¬ 등을 μœ„ν•œ νŽ˜μ΄μ§€μ™€ Rκ³Ό ν•˜λ‘‘μ„ 톡해 λΆ„μ„ν•œ ν†΅κ³„μžλ£Œλ₯Ό μœ„ν•œ νŽ˜μ΄μ§€ 등을 확인 ν•  수 μžˆμŠ΅λ‹ˆλ‹€. [Results & discussions] [ν•œκ³„] μ›Ήμ–΄ν”Œλ¦¬μΌ€μ΄μ…˜ κ΄€λ ¨ ν•˜μ—¬μ„œλŠ” μ‚¬μš©μž λ™μ‹œ 접속은 둜그인 μƒνƒœλ₯Ό μœ μ§€ν•˜κ³  μžˆλŠ” μ„Έμ…˜(session)의 갯수λ₯Ό μΈ‘μ •ν•˜λŠ” κ²ƒμž…λ‹ˆλ‹€. ν•˜μ§€λ§Œ μ›Ήμ„œλΉ„μŠ€λŠ” μ‚¬μš©μžκ°€ λ‘œκ·Έμ•„μ›ƒν•œ μ‹œμ μ„ νŒŒμ•…ν•˜κΈ°λŠ” μ–΄λ €μ› μœΌλ©° μ„œλ²„μ˜ 지원 상황(λ©”λͺ¨λ¦¬,CPU,λ„€νŠΈμ›Œν¬)에 따라 μ„±λŠ₯ 편차(였차)κ°€ λ°œμƒν•˜μ˜€μŠ΅λ‹ˆλ‹€.μˆœκ°„ μ²˜λ¦¬λŸ‰μœΌλ‘œ μ„œλ²„μ˜ λͺ¨λ“  μ„±λŠ₯을 평가할 수 μ—†λ‹€λŠ” 결둠이 λ‚˜μ™”μŠ΅λ‹ˆλ‹€. μ„±λŠ₯평가 μ‚¬μš©μž μž…μž₯μ—μ„œλŠ” μ„œλ²„ μ‘λ‹΅μ‹œκ°„μ΄ μ§§μ„μˆ˜λ‘ μ’‹μŒ μ„œλ²„κ°€ 아무리 λΉ λ₯΄λ”라도 μ„œλ²„μ™€ μ‚¬μš©μž μ‚¬μ΄μ—λŠ” λ„€νŠΈμ›Œν¬ νšŒμ„ μ΄ μ‘΄μž¬ν•˜κΈ° λ•Œλ¬Έμ— 지연 μ‹œκ°„(latency time)이 λ°œμƒν•  수 밖에 μ—†μŒ. μ„œλ²„ 섀계 및 κ΄€λ¦¬μž μž…μž₯μ—μ„œλŠ” μ„œλ²„ 쀑단(halt)κ°€ κ°€μž₯ 큰 λ¬Έμ œκ°€ 될 수 μžˆλ‹€. μ‚¬μš©μžμ˜ μ΅œμ ν•œ μ„œλΉ„μŠ€ κ²½ν—˜μ„ μœ„ν•œ "짧은 응닡 μ‹œκ°„(short response time)" κ³Ό μ„œλ²„μ˜ μ•ˆμ •μ μΈ μš΄μ˜μ„ μœ„ν•œ "μ μ ˆν•œ μ²˜λ¦¬λŸ‰(proper throughtput)"이 μ„±λŠ₯μ§€ν‘œμ˜ κ°€μž₯ 큰 기쀀이 λœλ‹€κ³  ν•  수 μžˆλ‹€. [μ—­λŸ‰] λ³Έ ν”„λ‘œμ νŠΈλ‘œλΆ€ν„° 빅데이터λ₯Ό μ΄μš©ν•œ λ‹€μ–‘ν•œ 뢄석 μ—­λŸ‰ 및 μ‹œμŠ€ν…œ κ΅¬μΆ•μ—­λŸ‰μ„ μŠ΅λ“ν•œ ν›„ κΈ°μ‘΄μ‹œμŠ€ν…œμ—μ„œ λ°œμƒν•œ λ‹€λŸ‰μ˜ 데이터λ₯Ό ν™œμš©ν•˜μ—¬ μ—¬λŸ¬ λΆ„μ•Όμ˜ λΉ„μ¦ˆλ‹ˆμŠ€κ°€ 진행될 수 μžˆμŒμ„ μ•Œκ²Œλ˜μ—ˆμŠ΅λ‹ˆλ‹€. λ¦¬λˆ…μŠ€ OS/κΈ°λ³Έμ§€μ‹μŠ΅λ“μœΌλ‘œλŠ” ν˜„μ—…μ—μ„œ μ• μš©ν•˜λŠ” λ¦¬λˆ…μŠ€λ₯Ό κ°œλ³„ PC에 μ„€μΉ˜/ν™œμš©ν•˜μ—¬ λŒ€ν˜•μ‹œμŠ€ν…œ 운영λŠ₯λ ₯을 μŠ΅λ“ν•˜κ³  λ¦¬λˆ…μŠ€ OS기반의 ν”„λ‘œκ·Έλž¨ 개발 λŠ₯λ ₯ 및 SQL μ‚¬μš© λŠ₯λ ₯을 μŠ΅λ“ν•˜μ˜€μŠ΅λ‹ˆλ‹€. 빅데이터 μ‹œμŠ€ν…œκ³Ό μ‘μš© μ• ν”Œλ¦¬μΌ€μ΄μ…˜ μ—°λ™μœΌλ‘œλŠ” 빅데이터 ν™˜κ²½μ—μ„œ λΆ„μ„λœ 자료λ₯Ό WebApplication 및 λ‹€μ–‘ν•œ Application 과의 연동을 톡해 뢄석 자료 Visualization κΈ°μˆ μ„ μŠ΅λ“ν•˜μ˜€μŠ΅λ‹ˆλ‹€. 빅데이터 μ‹œμŠ€ν…œ ꡬ좕/ λ‹€μ–‘ν•œ 빅데이터 뢄석 μ—­λŸ‰ μŠ΅λ“ν•˜μ˜€μŠ΅λ‹ˆλ‹€. 빅데이터 ν™˜κ²½μ„ μœ„ν•œ ν•˜λ‘‘ μ‹œμŠ€ν…œ ꡬ좕과 그에 μ—°κ΄€λœ ν•˜λ‘‘μ—μ½” μ‹œμŠ€ν…œμ„ ꡬ좕 ν•˜κ³  λ‹€μ–‘ν•œ ν˜„μ—…μ—μ„œ ν•„μš”λ‘œ ν•˜λŠ” 뢄석전문가가 되기 μœ„ν•΄ 빅데이터 뢄석 기법/μ „λž΅ μ‹€μŠ΅μ„ μ§„ν–‰ν•˜μ˜€μŠ΅λ‹ˆλ‹€. [Conclusion] λ³Έ ν”„λ‘œμ νŠΈμ™€ 같이 ν™ˆνŽ˜μ΄μ§€λ₯Ό κ°œμ„€, 운영 및 μœ μ§€ν•˜λŠ” κ³Όμ •κ³Ό λ”λΆˆμ–΄ μ‹€μ‹œλ‘œ λΆ„μ„ν•˜λŠ” 과정을 ν†΅ν•˜μ—¬ 톡합적이고 λŠκΉ€ μ—†λŠ” κ΄€λ¦¬μž-μ†ŒλΉ„μž 체인의 λΈŒλžœλ“œ 관계 ν˜•μ„±μ„ 보닀 μš©μ΄ν•˜κ²Œ ν•  수 μžˆλ‹€. μ™œλƒν•˜λ©΄ 운영자(κ΄€λ¦¬μž)λŠ” μ‹€μ‹œκ°„ λ°μ΄ν„°λ‘œκ·Έλ₯Ό ν†΅ν•˜μ—¬ μ†ŒλΉ„μžμ˜ ν™ˆνŽ˜μ΄μ§€μ—μ„œ ν–‰ν•˜λŠ” ν–‰μœ„λ“€μ„ λ°μ΄ν„°λ‘œ μ•Œ 수 있기 λ•Œλ¬Έμ΄λ‹€. μ—¬ν–‰μƒν’ˆμ΄ ꡬ글과 같은 ν¬ν„Έμ‚¬μ΄νŠΈμ˜ μ•Œκ³ λ¦¬μ¦˜μ— 잘 맞좰져 검색이 잘 λ˜λ„λ‘ λ…Έλ ₯ν•΄μ•Ό ν•˜λ©° μ½˜ν…μΈ  확산성을 μœ„ν•œ 연ꡬ 및 개발이 ν•„μš”ν•˜λ‹€. μ†ŒλΉ„μžλ“€μ€ 정보λ₯Ό κ²€μƒ‰ν•˜κ³ , κ΄€λ¦¬μžλŠ” μ†ŒλΉ„μžμ—κ²Œ 정보λ₯Ό μ œκ³΅ν•˜λŠ” λ°©μ‹μœΌλ‘œμ„œ 이 λ‘˜μ„ λ§€ν•‘ν•˜λŠ” λ…Έλ ₯이 ν•„μš”ν•˜κ²Œ 될 것이며 λΉ„μ •ν˜•ν™”λœ 데이터λ₯Ό μ΄μš©κ°€λŠ₯ν•œ μ •ν˜•λ°μ΄ν„°λ‘œ λ§Œλ“œλŠ” μž‘μ—…μ„ μš©μ΄ν•˜κ²Œ ν•˜μ—¬ λ”œλ ˆλ§ˆλ₯Ό μ΅œμ†Œν™”ν•˜μ—¬ ν•œλ‹€.]]>
kor ver. [Abstract] κ³ κ°λ§žμΆ€ν˜• μ—¬ν–‰μ‚¬μ΄νŠΈοΌˆFront-Back-End / Bigdata μ—­λŸ‰οΌ‰ 아킀텍쳐 : Linux（CentOsοΌ‰ -＞ Java -＞ JSP -＞ Mysql-＞ Hadoop -＞ R -＞ Output λͺ©μ  - μ—¬ν–‰μƒν’ˆμ„ κ΄€λ¦¬μžμ—κ²Œ μ‚¬μš©μžκ°€ μ›ν•˜λŠ” 정보λ₯Ό μž…λ ₯ν•˜κ³  μ‹ μ²­ν•˜μ—¬ κ·Έ μ •λ³΄οΌˆλ°μ΄ν„°οΌ‰λ₯Ό κ΄€λ¦¬μžκ°€ 그에 λ§žλŠ” 여행사 νŒ¨ν‚€μ§€μƒν’ˆμ„ 제곡 ν•˜λ„λ‘ ν™ˆνŽ˜μ΄μ§€λ₯Ό λ§Œλ“€κ³  λ‘œκ·Έλ°μ΄ν„°λ‘œ μ†ŒλΉ„μžμ„±ν–₯을 λΆ„μ„ν•˜λŠ” ν”„λ‘œμ νŠΈ μ˜€μŠ΅λ‹ˆλ‹€. 1. 고객 λ§žμΆ€ν˜• μ—¬ν–‰μΆ”μ²œ μ‚¬μ΄νŠΈ μ œμž‘ 2. λ°˜μ‘ν˜• μ›Ήμ‚¬μ΄νŠΈ κ΅¬ν˜„ 3. λ°μ΄ν„°λ² μ΄μŠ€ ν™œμš© 및 λΉ„μ •ν˜• 데이터(μ•‘μ„ΈμŠ€ 둜그) 뢄석 4. λΆ„μ‚°νŒŒμΌμ‹œμŠ€ν…œ Hadoop ν™œμš© 5. μ‹€μ‹œκ°„ 톡계 및 μΈμ‚¬μ΄νŠΈ λ„μΆœ [Method & Process] λ³Έ ν”„λ‘œμ νŠΈλŠ” μ‚¬μš©μž μ€‘μ‹¬μ˜ tour ν™ˆνŽ˜μ΄μ§€λ₯Ό λ§Œλ“€μ–΄ μ—¬λŸ¬ μ†Œν”„νŠΈμ›¨μ–΄ 툴과 κΈ°λ°˜μ„ μ‘μš©ν•˜μ—¬ 섀계뢀터 μ‹€ν–‰λ‹¨κ³„κΉŒμ§€μ˜ 과정을 거쳐 ν”„λ‘œμ νŠΈλ₯Ό μ§„ν–‰ν•˜μ˜€μŠ΅λ‹ˆλ‹€. Customer-Centered Tour λΌλŠ” 주제λ₯Ό μ •ν•˜μ—¬ μ—¬ν–‰μƒν’ˆμ„ κ΄€λ¦¬μžμ—κ²Œ μ‚¬μš©μžκ°€ μ›ν•˜λŠ” 정보λ₯Ό μž…λ ₯ν•˜κ³  μ‹ μ²­ν•˜μ—¬ κ·Έ 정보(데이터)λ₯Ό κ΄€λ¦¬μžκ°€ 그에 λ§žλŠ” 여행사 νŒ¨ν‚€μ§€μƒν’ˆμ„ 제곡 ν•˜λ„λ‘ μ„€κ³„ν•˜μ˜€μŠ΅λ‹ˆλ‹€. MVCλͺ¨λΈμ„ 기점으둜 ν™˜κ²½μ„€μ • ν”„λ‘œκ·Έλž¨, μ†Œν”„νŠΈμ›¨μ–΄, 뢄석틀을 μ΄μš©ν•˜μ—¬ ν”„λ‘œμ νŠΈλ₯Ό 연ꡬ 및 μ§„ν–‰ν•˜μ˜€μœΌλ©°, Customer-Based Tour ν™ˆνŽ˜μ΄μ§€λ₯Ό μ„€κ³„ν•˜μ—¬ μ†Œν”„νŠΈμ›¨μ–΄ μ€‘μ‹¬μœΌλ‘œ JAVA와 같은 객체지ν–₯언어와 κ΄€κ³„ν˜• λ°μ΄ν„°λ² μ΄μŠ€(Oracle11g)의 ν™˜κ²½μ„€μ •μ΄λ‚˜ (Linux) μ†Œν”„νŠΈμ›¨μ–΄λ₯Ό μ—°λ™ν•˜μ—¬ λΆ„μ„ν‹€λ‘œ 데이터λ₯Ό λΆ„μ„ν•˜λŠ” ν”„λ‘œκ·Έλž˜λ°(R, ν•˜λ‘‘)을 μ΄μš©ν•˜μ—¬ μž…μΆœλ ₯ 된 데이터λ₯Ό λΆ„μ„ν•˜μ—¬ ν”„λ‘œμ νŠΈλ₯Ό μ™„λ£Œν•˜μ˜€μŠ΅λ‹ˆ. 각 ν”„λ‘œκ·Έλž¨κ³Ό μ• ν”Œλ¦¬μΌ€μ΄μ…˜νˆ΄μ„ μ†Œκ°œν•˜κ³  μ‘μš© 및 μ—°κ΅¬μ§„ν–‰μ ˆμ°¨λ₯Ό μž‘μ„±ν•˜μ˜€μŠ΅λ‹ˆλ‹€. Customer Mode μ‚¬μš©μžκ°€ 메인 νŽ˜μ΄μ§€(index)λ₯Ό 기점으둜 νšŒμ›κ°€μž… νŽ˜μ΄μ§€λ₯Ό 톡해 아이디λ₯Ό λ§Œλ“€μ–΄ νšŒμ›μΈμ¦μ„ ν•˜μ—¬ Customer mode둜 μ ‘μ†ν•©λ‹ˆλ‹€. Customer Modeμ—μ„œλŠ” ν™ˆνŽ˜μ΄μ§€ μƒμœ„μ— 더 λ§Žμ€ μΉ΄ν…Œκ³ λ¦¬λ₯Ό 찾을 수 μžˆμŠ΅λ‹ˆλ‹€. λ˜ν•œ 메인 νŽ˜μ΄μ§€ ν•˜λ‹¨μ—μ„œ μ‚¬μš©μžκ°€ μ›ν•˜λŠ” 각 λ‚˜λΌλ³„ 여행지λ₯Ό ν΄λ¦­ν•˜μ—¬ λ‹€μŒνŽ˜μ΄μ§€μΈ λ‚˜λΌλ³„μƒμ„Έμ •λ³΄λ₯Ό λ³Ό 수 있음과 λ™μ‹œμ— νŒ¨ν‚€μ§€μƒν’ˆμ„ μ‹ μ²­ν•  수 μžˆλŠ” νŽ˜μ΄μ§€λ‘œ 가도둝 링크λ₯Ό μ„€κ³„ν•˜μ˜€μŠ΅λ‹ˆλ‹€. λ˜ν•œ μ‚¬μš©μžλŠ” ν™ˆνŽ˜μ΄μ§€μ—μ„œ μ œκ³΅ν•˜λŠ” 여행지 톡계λ₯Ό 각 λ‚˜λΌλ³„, ν…Œλ§ˆλ³„, 계쑀별, μ—°λ Ήλ³„λ‘œ λ³Ό 수 μžˆλ„λ‘ μ œκ³΅λ©λ‹ˆλ‹€. Administrator Mode νŒ¨ν‚€μ§€μƒν’ˆμ„ λ“±λ‘ν•˜λ©΄ κ·Έ 정보가 κ΄€λ¦¬μž(Administrator)μ—κ²Œ λ‘œκ·Έλ°μ΄ν„°κ°€ μž…λ ₯되고 여행사별 νŒ¨ν‚€μ§€μƒν’ˆκ³Ό μ‚¬μš©μž(Customer Mode)μ—κ²Œ 연결을 ν•˜μ—¬ λ‹€μ‹œ μ‚¬μš©μžμ—κ²Œ λ³΄λ‚΄μ€λ‹ˆλ‹€. κ΄€λ¦¬μžλŠ” VIP고객정보, νšŒμ›κ΄€λ¦¬ 등을 μœ„ν•œ νŽ˜μ΄μ§€μ™€ Rκ³Ό ν•˜λ‘‘μ„ 톡해 λΆ„μ„ν•œ ν†΅κ³„μžλ£Œλ₯Ό μœ„ν•œ νŽ˜μ΄μ§€ 등을 확인 ν•  수 μžˆμŠ΅λ‹ˆλ‹€. [Results & discussions] [ν•œκ³„] μ›Ήμ–΄ν”Œλ¦¬μΌ€μ΄μ…˜ κ΄€λ ¨ ν•˜μ—¬μ„œλŠ” μ‚¬μš©μž λ™μ‹œ 접속은 둜그인 μƒνƒœλ₯Ό μœ μ§€ν•˜κ³  μžˆλŠ” μ„Έμ…˜(session)의 갯수λ₯Ό μΈ‘μ •ν•˜λŠ” κ²ƒμž…λ‹ˆλ‹€. ν•˜μ§€λ§Œ μ›Ήμ„œλΉ„μŠ€λŠ” μ‚¬μš©μžκ°€ λ‘œκ·Έμ•„μ›ƒν•œ μ‹œμ μ„ νŒŒμ•…ν•˜κΈ°λŠ” μ–΄λ €μ› μœΌλ©° μ„œλ²„μ˜ 지원 상황(λ©”λͺ¨λ¦¬,CPU,λ„€νŠΈμ›Œν¬)에 따라 μ„±λŠ₯ 편차(였차)κ°€ λ°œμƒν•˜μ˜€μŠ΅λ‹ˆλ‹€.μˆœκ°„ μ²˜λ¦¬λŸ‰μœΌλ‘œ μ„œλ²„μ˜ λͺ¨λ“  μ„±λŠ₯을 평가할 수 μ—†λ‹€λŠ” 결둠이 λ‚˜μ™”μŠ΅λ‹ˆλ‹€. μ„±λŠ₯평가 μ‚¬μš©μž μž…μž₯μ—μ„œλŠ” μ„œλ²„ μ‘λ‹΅μ‹œκ°„μ΄ μ§§μ„μˆ˜λ‘ μ’‹μŒ μ„œλ²„κ°€ 아무리 λΉ λ₯΄λ”라도 μ„œλ²„μ™€ μ‚¬μš©μž μ‚¬μ΄μ—λŠ” λ„€νŠΈμ›Œν¬ νšŒμ„ μ΄ μ‘΄μž¬ν•˜κΈ° λ•Œλ¬Έμ— 지연 μ‹œκ°„(latency time)이 λ°œμƒν•  수 밖에 μ—†μŒ. μ„œλ²„ 섀계 및 κ΄€λ¦¬μž μž…μž₯μ—μ„œλŠ” μ„œλ²„ 쀑단(halt)κ°€ κ°€μž₯ 큰 λ¬Έμ œκ°€ 될 수 μžˆλ‹€. μ‚¬μš©μžμ˜ μ΅œμ ν•œ μ„œλΉ„μŠ€ κ²½ν—˜μ„ μœ„ν•œ "짧은 응닡 μ‹œκ°„(short response time)" κ³Ό μ„œλ²„μ˜ μ•ˆμ •μ μΈ μš΄μ˜μ„ μœ„ν•œ "μ μ ˆν•œ μ²˜λ¦¬λŸ‰(proper throughtput)"이 μ„±λŠ₯μ§€ν‘œμ˜ κ°€μž₯ 큰 기쀀이 λœλ‹€κ³  ν•  수 μžˆλ‹€. [μ—­λŸ‰] λ³Έ ν”„λ‘œμ νŠΈλ‘œλΆ€ν„° 빅데이터λ₯Ό μ΄μš©ν•œ λ‹€μ–‘ν•œ 뢄석 μ—­λŸ‰ 및 μ‹œμŠ€ν…œ κ΅¬μΆ•μ—­λŸ‰μ„ μŠ΅λ“ν•œ ν›„ κΈ°μ‘΄μ‹œμŠ€ν…œμ—μ„œ λ°œμƒν•œ λ‹€λŸ‰μ˜ 데이터λ₯Ό ν™œμš©ν•˜μ—¬ μ—¬λŸ¬ λΆ„μ•Όμ˜ λΉ„μ¦ˆλ‹ˆμŠ€κ°€ 진행될 수 μžˆμŒμ„ μ•Œκ²Œλ˜μ—ˆμŠ΅λ‹ˆλ‹€. λ¦¬λˆ…μŠ€ OS/κΈ°λ³Έμ§€μ‹μŠ΅λ“μœΌλ‘œλŠ” ν˜„μ—…μ—μ„œ μ• μš©ν•˜λŠ” λ¦¬λˆ…μŠ€λ₯Ό κ°œλ³„ PC에 μ„€μΉ˜/ν™œμš©ν•˜μ—¬ λŒ€ν˜•μ‹œμŠ€ν…œ 운영λŠ₯λ ₯을 μŠ΅λ“ν•˜κ³  λ¦¬λˆ…μŠ€ OS기반의 ν”„λ‘œκ·Έλž¨ 개발 λŠ₯λ ₯ 및 SQL μ‚¬μš© λŠ₯λ ₯을 μŠ΅λ“ν•˜μ˜€μŠ΅λ‹ˆλ‹€. 빅데이터 μ‹œμŠ€ν…œκ³Ό μ‘μš© μ• ν”Œλ¦¬μΌ€μ΄μ…˜ μ—°λ™μœΌλ‘œλŠ” 빅데이터 ν™˜κ²½μ—μ„œ λΆ„μ„λœ 자료λ₯Ό WebApplication 및 λ‹€μ–‘ν•œ Application 과의 연동을 톡해 뢄석 자료 Visualization κΈ°μˆ μ„ μŠ΅λ“ν•˜μ˜€μŠ΅λ‹ˆλ‹€. 빅데이터 μ‹œμŠ€ν…œ ꡬ좕/ λ‹€μ–‘ν•œ 빅데이터 뢄석 μ—­λŸ‰ μŠ΅λ“ν•˜μ˜€μŠ΅λ‹ˆλ‹€. 빅데이터 ν™˜κ²½μ„ μœ„ν•œ ν•˜λ‘‘ μ‹œμŠ€ν…œ ꡬ좕과 그에 μ—°κ΄€λœ ν•˜λ‘‘μ—μ½” μ‹œμŠ€ν…œμ„ ꡬ좕 ν•˜κ³  λ‹€μ–‘ν•œ ν˜„μ—…μ—μ„œ ν•„μš”λ‘œ ν•˜λŠ” 뢄석전문가가 되기 μœ„ν•΄ 빅데이터 뢄석 기법/μ „λž΅ μ‹€μŠ΅μ„ μ§„ν–‰ν•˜μ˜€μŠ΅λ‹ˆλ‹€. [Conclusion] λ³Έ ν”„λ‘œμ νŠΈμ™€ 같이 ν™ˆνŽ˜μ΄μ§€λ₯Ό κ°œμ„€, 운영 및 μœ μ§€ν•˜λŠ” κ³Όμ •κ³Ό λ”λΆˆμ–΄ μ‹€μ‹œλ‘œ λΆ„μ„ν•˜λŠ” 과정을 ν†΅ν•˜μ—¬ 톡합적이고 λŠκΉ€ μ—†λŠ” κ΄€λ¦¬μž-μ†ŒλΉ„μž 체인의 λΈŒλžœλ“œ 관계 ν˜•μ„±μ„ 보닀 μš©μ΄ν•˜κ²Œ ν•  수 μžˆλ‹€. μ™œλƒν•˜λ©΄ 운영자(κ΄€λ¦¬μž)λŠ” μ‹€μ‹œκ°„ λ°μ΄ν„°λ‘œκ·Έλ₯Ό ν†΅ν•˜μ—¬ μ†ŒλΉ„μžμ˜ ν™ˆνŽ˜μ΄μ§€μ—μ„œ ν–‰ν•˜λŠ” ν–‰μœ„λ“€μ„ λ°μ΄ν„°λ‘œ μ•Œ 수 있기 λ•Œλ¬Έμ΄λ‹€. μ—¬ν–‰μƒν’ˆμ΄ ꡬ글과 같은 ν¬ν„Έμ‚¬μ΄νŠΈμ˜ μ•Œκ³ λ¦¬μ¦˜μ— 잘 맞좰져 검색이 잘 λ˜λ„λ‘ λ…Έλ ₯ν•΄μ•Ό ν•˜λ©° μ½˜ν…μΈ  확산성을 μœ„ν•œ 연ꡬ 및 개발이 ν•„μš”ν•˜λ‹€. μ†ŒλΉ„μžλ“€μ€ 정보λ₯Ό κ²€μƒ‰ν•˜κ³ , κ΄€λ¦¬μžλŠ” μ†ŒλΉ„μžμ—κ²Œ 정보λ₯Ό μ œκ³΅ν•˜λŠ” λ°©μ‹μœΌλ‘œμ„œ 이 λ‘˜μ„ λ§€ν•‘ν•˜λŠ” λ…Έλ ₯이 ν•„μš”ν•˜κ²Œ 될 것이며 λΉ„μ •ν˜•ν™”λœ 데이터λ₯Ό μ΄μš©κ°€λŠ₯ν•œ μ •ν˜•λ°μ΄ν„°λ‘œ λ§Œλ“œλŠ” μž‘μ—…μ„ μš©μ΄ν•˜κ²Œ ν•˜μ—¬ λ”œλ ˆλ§ˆλ₯Ό μ΅œμ†Œν™”ν•˜μ—¬ ν•œλ‹€.]]> Wed, 27 Dec 2017 15:57:14 GMT /ssuserf03d2b/kor-verglobal-go-bigdatacloud-computing-project-mainly-in-mvc-model2 ssuserf03d2b@slideshare.net(ssuserf03d2b) [kor ver.]Global GO (Bigdata-Cloud computing project - mainly in MVC model2) ssuserf03d2b kor ver. [Abstract] κ³ κ°λ§žμΆ€ν˜• μ—¬ν–‰μ‚¬μ΄νŠΈοΌˆFront-Back-End / Bigdata μ—­λŸ‰οΌ‰ 아킀텍쳐 : Linux（CentOsοΌ‰ -＞ Java -＞ JSP -＞ Mysql-＞ Hadoop -＞ R -＞ Output λͺ©μ  - μ—¬ν–‰μƒν’ˆμ„ κ΄€λ¦¬μžμ—κ²Œ μ‚¬μš©μžκ°€ μ›ν•˜λŠ” 정보λ₯Ό μž…λ ₯ν•˜κ³  μ‹ μ²­ν•˜μ—¬ κ·Έ μ •λ³΄οΌˆλ°μ΄ν„°οΌ‰λ₯Ό κ΄€λ¦¬μžκ°€ 그에 λ§žλŠ” 여행사 νŒ¨ν‚€μ§€μƒν’ˆμ„ 제곡 ν•˜λ„λ‘ ν™ˆνŽ˜μ΄μ§€λ₯Ό λ§Œλ“€κ³  λ‘œκ·Έλ°μ΄ν„°λ‘œ μ†ŒλΉ„μžμ„±ν–₯을 λΆ„μ„ν•˜λŠ” ν”„λ‘œμ νŠΈ μ˜€μŠ΅λ‹ˆλ‹€. 1. 고객 λ§žμΆ€ν˜• μ—¬ν–‰μΆ”μ²œ μ‚¬μ΄νŠΈ μ œμž‘ 2. λ°˜μ‘ν˜• μ›Ήμ‚¬μ΄νŠΈ κ΅¬ν˜„ 3. λ°μ΄ν„°λ² μ΄μŠ€ ν™œμš© 및 λΉ„μ •ν˜• 데이터(μ•‘μ„ΈμŠ€ 둜그) 뢄석 4. λΆ„μ‚°νŒŒμΌμ‹œμŠ€ν…œ Hadoop ν™œμš© 5. μ‹€μ‹œκ°„ 톡계 및 μΈμ‚¬μ΄νŠΈ λ„μΆœ [Method & Process] λ³Έ ν”„λ‘œμ νŠΈλŠ” μ‚¬μš©μž μ€‘μ‹¬μ˜ tour ν™ˆνŽ˜μ΄μ§€λ₯Ό λ§Œλ“€μ–΄ μ—¬λŸ¬ μ†Œν”„νŠΈμ›¨μ–΄ 툴과 κΈ°λ°˜μ„ μ‘μš©ν•˜μ—¬ 섀계뢀터 μ‹€ν–‰λ‹¨κ³„κΉŒμ§€μ˜ 과정을 거쳐 ν”„λ‘œμ νŠΈλ₯Ό μ§„ν–‰ν•˜μ˜€μŠ΅λ‹ˆλ‹€. Customer-Centered Tour λΌλŠ” 주제λ₯Ό μ •ν•˜μ—¬ μ—¬ν–‰μƒν’ˆμ„ κ΄€λ¦¬μžμ—κ²Œ μ‚¬μš©μžκ°€ μ›ν•˜λŠ” 정보λ₯Ό μž…λ ₯ν•˜κ³  μ‹ μ²­ν•˜μ—¬ κ·Έ 정보(데이터)λ₯Ό κ΄€λ¦¬μžκ°€ 그에 λ§žλŠ” 여행사 νŒ¨ν‚€μ§€μƒν’ˆμ„ 제곡 ν•˜λ„λ‘ μ„€κ³„ν•˜μ˜€μŠ΅λ‹ˆλ‹€. MVCλͺ¨λΈμ„ 기점으둜 ν™˜κ²½μ„€μ • ν”„λ‘œκ·Έλž¨, μ†Œν”„νŠΈμ›¨μ–΄, 뢄석틀을 μ΄μš©ν•˜μ—¬ ν”„λ‘œμ νŠΈλ₯Ό 연ꡬ 및 μ§„ν–‰ν•˜μ˜€μœΌλ©°, Customer-Based Tour ν™ˆνŽ˜μ΄μ§€λ₯Ό μ„€κ³„ν•˜μ—¬ μ†Œν”„νŠΈμ›¨μ–΄ μ€‘μ‹¬μœΌλ‘œ JAVA와 같은 객체지ν–₯언어와 κ΄€κ³„ν˜• λ°μ΄ν„°λ² μ΄μŠ€(Oracle11g)의 ν™˜κ²½μ„€μ •μ΄λ‚˜ (Linux) μ†Œν”„νŠΈμ›¨μ–΄λ₯Ό μ—°λ™ν•˜μ—¬ λΆ„μ„ν‹€λ‘œ 데이터λ₯Ό λΆ„μ„ν•˜λŠ” ν”„λ‘œκ·Έλž˜λ°(R, ν•˜λ‘‘)을 μ΄μš©ν•˜μ—¬ μž…μΆœλ ₯ 된 데이터λ₯Ό λΆ„μ„ν•˜μ—¬ ν”„λ‘œμ νŠΈλ₯Ό μ™„λ£Œν•˜μ˜€μŠ΅λ‹ˆ. 각 ν”„λ‘œκ·Έλž¨κ³Ό μ• ν”Œλ¦¬μΌ€μ΄μ…˜νˆ΄μ„ μ†Œκ°œν•˜κ³  μ‘μš© 및 μ—°κ΅¬μ§„ν–‰μ ˆμ°¨λ₯Ό μž‘μ„±ν•˜μ˜€μŠ΅λ‹ˆλ‹€. Customer Mode μ‚¬μš©μžκ°€ 메인 νŽ˜μ΄μ§€(index)λ₯Ό 기점으둜 νšŒμ›κ°€μž… νŽ˜μ΄μ§€λ₯Ό 톡해 아이디λ₯Ό λ§Œλ“€μ–΄ νšŒμ›μΈμ¦μ„ ν•˜μ—¬ Customer mode둜 μ ‘μ†ν•©λ‹ˆλ‹€. Customer Modeμ—μ„œλŠ” ν™ˆνŽ˜μ΄μ§€ μƒμœ„μ— 더 λ§Žμ€ μΉ΄ν…Œκ³ λ¦¬λ₯Ό 찾을 수 μžˆμŠ΅λ‹ˆλ‹€. λ˜ν•œ 메인 νŽ˜μ΄μ§€ ν•˜λ‹¨μ—μ„œ μ‚¬μš©μžκ°€ μ›ν•˜λŠ” 각 λ‚˜λΌλ³„ 여행지λ₯Ό ν΄λ¦­ν•˜μ—¬ λ‹€μŒνŽ˜μ΄μ§€μΈ λ‚˜λΌλ³„μƒμ„Έμ •λ³΄λ₯Ό λ³Ό 수 있음과 λ™μ‹œμ— νŒ¨ν‚€μ§€μƒν’ˆμ„ μ‹ μ²­ν•  수 μžˆλŠ” νŽ˜μ΄μ§€λ‘œ 가도둝 링크λ₯Ό μ„€κ³„ν•˜μ˜€μŠ΅λ‹ˆλ‹€. λ˜ν•œ μ‚¬μš©μžλŠ” ν™ˆνŽ˜μ΄μ§€μ—μ„œ μ œκ³΅ν•˜λŠ” 여행지 톡계λ₯Ό 각 λ‚˜λΌλ³„, ν…Œλ§ˆλ³„, 계쑀별, μ—°λ Ήλ³„λ‘œ λ³Ό 수 μžˆλ„λ‘ μ œκ³΅λ©λ‹ˆλ‹€. Administrator Mode νŒ¨ν‚€μ§€μƒν’ˆμ„ λ“±λ‘ν•˜λ©΄ κ·Έ 정보가 κ΄€λ¦¬μž(Administrator)μ—κ²Œ λ‘œκ·Έλ°μ΄ν„°κ°€ μž…λ ₯되고 여행사별 νŒ¨ν‚€μ§€μƒν’ˆκ³Ό μ‚¬μš©μž(Customer Mode)μ—κ²Œ 연결을 ν•˜μ—¬ λ‹€μ‹œ μ‚¬μš©μžμ—κ²Œ λ³΄λ‚΄μ€λ‹ˆλ‹€. κ΄€λ¦¬μžλŠ” VIP고객정보, νšŒμ›κ΄€λ¦¬ 등을 μœ„ν•œ νŽ˜μ΄μ§€μ™€ Rκ³Ό ν•˜λ‘‘μ„ 톡해 λΆ„μ„ν•œ ν†΅κ³„μžλ£Œλ₯Ό μœ„ν•œ νŽ˜μ΄μ§€ 등을 확인 ν•  수 μžˆμŠ΅λ‹ˆλ‹€. [Results & discussions] [ν•œκ³„] μ›Ήμ–΄ν”Œλ¦¬μΌ€μ΄μ…˜ κ΄€λ ¨ ν•˜μ—¬μ„œλŠ” μ‚¬μš©μž λ™μ‹œ 접속은 둜그인 μƒνƒœλ₯Ό μœ μ§€ν•˜κ³  μžˆλŠ” μ„Έμ…˜(session)의 갯수λ₯Ό μΈ‘μ •ν•˜λŠ” κ²ƒμž…λ‹ˆλ‹€. ν•˜μ§€λ§Œ μ›Ήμ„œλΉ„μŠ€λŠ” μ‚¬μš©μžκ°€ λ‘œκ·Έμ•„μ›ƒν•œ μ‹œμ μ„ νŒŒμ•…ν•˜κΈ°λŠ” μ–΄λ €μ› μœΌλ©° μ„œλ²„μ˜ 지원 상황(λ©”λͺ¨λ¦¬,CPU,λ„€νŠΈμ›Œν¬)에 따라 μ„±λŠ₯ 편차(였차)κ°€ λ°œμƒν•˜μ˜€μŠ΅λ‹ˆλ‹€.μˆœκ°„ μ²˜λ¦¬λŸ‰μœΌλ‘œ μ„œλ²„μ˜ λͺ¨λ“  μ„±λŠ₯을 평가할 수 μ—†λ‹€λŠ” 결둠이 λ‚˜μ™”μŠ΅λ‹ˆλ‹€. μ„±λŠ₯평가 μ‚¬μš©μž μž…μž₯μ—μ„œλŠ” μ„œλ²„ μ‘λ‹΅μ‹œκ°„μ΄ μ§§μ„μˆ˜λ‘ μ’‹μŒ μ„œλ²„κ°€ 아무리 λΉ λ₯΄λ”라도 μ„œλ²„μ™€ μ‚¬μš©μž μ‚¬μ΄μ—λŠ” λ„€νŠΈμ›Œν¬ νšŒμ„ μ΄ μ‘΄μž¬ν•˜κΈ° λ•Œλ¬Έμ— 지연 μ‹œκ°„(latency time)이 λ°œμƒν•  수 밖에 μ—†μŒ. μ„œλ²„ 섀계 및 κ΄€λ¦¬μž μž…μž₯μ—μ„œλŠ” μ„œλ²„ 쀑단(halt)κ°€ κ°€μž₯ 큰 λ¬Έμ œκ°€ 될 수 μžˆλ‹€. μ‚¬μš©μžμ˜ μ΅œμ ν•œ μ„œλΉ„μŠ€ κ²½ν—˜μ„ μœ„ν•œ "짧은 응닡 μ‹œκ°„(short response time)" κ³Ό μ„œλ²„μ˜ μ•ˆμ •μ μΈ μš΄μ˜μ„ μœ„ν•œ "μ μ ˆν•œ μ²˜λ¦¬λŸ‰(proper throughtput)"이 μ„±λŠ₯μ§€ν‘œμ˜ κ°€μž₯ 큰 기쀀이 λœλ‹€κ³  ν•  수 μžˆλ‹€. [μ—­λŸ‰] λ³Έ ν”„λ‘œμ νŠΈλ‘œλΆ€ν„° 빅데이터λ₯Ό μ΄μš©ν•œ λ‹€μ–‘ν•œ 뢄석 μ—­λŸ‰ 및 μ‹œμŠ€ν…œ κ΅¬μΆ•μ—­λŸ‰μ„ μŠ΅λ“ν•œ ν›„ κΈ°μ‘΄μ‹œμŠ€ν…œμ—μ„œ λ°œμƒν•œ λ‹€λŸ‰μ˜ 데이터λ₯Ό ν™œμš©ν•˜μ—¬ μ—¬λŸ¬ λΆ„μ•Όμ˜ λΉ„μ¦ˆλ‹ˆμŠ€κ°€ 진행될 수 μžˆμŒμ„ μ•Œκ²Œλ˜μ—ˆμŠ΅λ‹ˆλ‹€. λ¦¬λˆ…μŠ€ OS/κΈ°λ³Έμ§€μ‹μŠ΅λ“μœΌλ‘œλŠ” ν˜„μ—…μ—μ„œ μ• μš©ν•˜λŠ” λ¦¬λˆ…μŠ€λ₯Ό κ°œλ³„ PC에 μ„€μΉ˜/ν™œμš©ν•˜μ—¬ λŒ€ν˜•μ‹œμŠ€ν…œ 운영λŠ₯λ ₯을 μŠ΅λ“ν•˜κ³  λ¦¬λˆ…μŠ€ OS기반의 ν”„λ‘œκ·Έλž¨ 개발 λŠ₯λ ₯ 및 SQL μ‚¬μš© λŠ₯λ ₯을 μŠ΅λ“ν•˜μ˜€μŠ΅λ‹ˆλ‹€. 빅데이터 μ‹œμŠ€ν…œκ³Ό μ‘μš© μ• ν”Œλ¦¬μΌ€μ΄μ…˜ μ—°λ™μœΌλ‘œλŠ” 빅데이터 ν™˜κ²½μ—μ„œ λΆ„μ„λœ 자료λ₯Ό WebApplication 및 λ‹€μ–‘ν•œ Application 과의 연동을 톡해 뢄석 자료 Visualization κΈ°μˆ μ„ μŠ΅λ“ν•˜μ˜€μŠ΅λ‹ˆλ‹€. 빅데이터 μ‹œμŠ€ν…œ ꡬ좕/ λ‹€μ–‘ν•œ 빅데이터 뢄석 μ—­λŸ‰ μŠ΅λ“ν•˜μ˜€μŠ΅λ‹ˆλ‹€. 빅데이터 ν™˜κ²½μ„ μœ„ν•œ ν•˜λ‘‘ μ‹œμŠ€ν…œ ꡬ좕과 그에 μ—°κ΄€λœ ν•˜λ‘‘μ—μ½” μ‹œμŠ€ν…œμ„ ꡬ좕 ν•˜κ³  λ‹€μ–‘ν•œ ν˜„μ—…μ—μ„œ ν•„μš”λ‘œ ν•˜λŠ” 뢄석전문가가 되기 μœ„ν•΄ 빅데이터 뢄석 기법/μ „λž΅ μ‹€μŠ΅μ„ μ§„ν–‰ν•˜μ˜€μŠ΅λ‹ˆλ‹€. [Conclusion] λ³Έ ν”„λ‘œμ νŠΈμ™€ 같이 ν™ˆνŽ˜μ΄μ§€λ₯Ό κ°œμ„€, 운영 및 μœ μ§€ν•˜λŠ” κ³Όμ •κ³Ό λ”λΆˆμ–΄ μ‹€μ‹œλ‘œ λΆ„μ„ν•˜λŠ” 과정을 ν†΅ν•˜μ—¬ 톡합적이고 λŠκΉ€ μ—†λŠ” κ΄€λ¦¬μž-μ†ŒλΉ„μž 체인의 λΈŒλžœλ“œ 관계 ν˜•μ„±μ„ 보닀 μš©μ΄ν•˜κ²Œ ν•  수 μžˆλ‹€. μ™œλƒν•˜λ©΄ 운영자(κ΄€λ¦¬μž)λŠ” μ‹€μ‹œκ°„ λ°μ΄ν„°λ‘œκ·Έλ₯Ό ν†΅ν•˜μ—¬ μ†ŒλΉ„μžμ˜ ν™ˆνŽ˜μ΄μ§€μ—μ„œ ν–‰ν•˜λŠ” ν–‰μœ„λ“€μ„ λ°μ΄ν„°λ‘œ μ•Œ 수 있기 λ•Œλ¬Έμ΄λ‹€. μ—¬ν–‰μƒν’ˆμ΄ ꡬ글과 같은 ν¬ν„Έμ‚¬μ΄νŠΈμ˜ μ•Œκ³ λ¦¬μ¦˜μ— 잘 맞좰져 검색이 잘 λ˜λ„λ‘ λ…Έλ ₯ν•΄μ•Ό ν•˜λ©° μ½˜ν…μΈ  확산성을 μœ„ν•œ 연ꡬ 및 개발이 ν•„μš”ν•˜λ‹€. μ†ŒλΉ„μžλ“€μ€ 정보λ₯Ό κ²€μƒ‰ν•˜κ³ , κ΄€λ¦¬μžλŠ” μ†ŒλΉ„μžμ—κ²Œ 정보λ₯Ό μ œκ³΅ν•˜λŠ” λ°©μ‹μœΌλ‘œμ„œ 이 λ‘˜μ„ λ§€ν•‘ν•˜λŠ” λ…Έλ ₯이 ν•„μš”ν•˜κ²Œ 될 것이며 λΉ„μ •ν˜•ν™”λœ 데이터λ₯Ό μ΄μš©κ°€λŠ₯ν•œ μ •ν˜•λ°μ΄ν„°λ‘œ λ§Œλ“œλŠ” μž‘μ—…μ„ μš©μ΄ν•˜κ²Œ ν•˜μ—¬ λ”œλ ˆλ§ˆλ₯Ό μ΅œμ†Œν™”ν•˜μ—¬ ν•œλ‹€. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/random-171227155714-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> kor ver. [Abstract] κ³ κ°λ§žμΆ€ν˜• μ—¬ν–‰μ‚¬μ΄νŠΈοΌˆFront-Back-End / Bigdata μ—­λŸ‰οΌ‰ 아킀텍쳐 : Linux（CentOsοΌ‰ -＞ Java -＞ JSP -＞ Mysql-＞ Hadoop -＞ R -＞ Output λͺ©μ  - μ—¬ν–‰μƒν’ˆμ„ κ΄€λ¦¬μžμ—κ²Œ μ‚¬μš©μžκ°€ μ›ν•˜λŠ” 정보λ₯Ό μž…λ ₯ν•˜κ³  μ‹ μ²­ν•˜μ—¬ κ·Έ μ •λ³΄οΌˆλ°μ΄ν„°οΌ‰λ₯Ό κ΄€λ¦¬μžκ°€ 그에 λ§žλŠ” 여행사 νŒ¨ν‚€μ§€μƒν’ˆμ„ 제곡 ν•˜λ„λ‘ ν™ˆνŽ˜μ΄μ§€λ₯Ό λ§Œλ“€κ³  λ‘œκ·Έλ°μ΄ν„°λ‘œ μ†ŒλΉ„μžμ„±ν–₯을 λΆ„μ„ν•˜λŠ” ν”„λ‘œμ νŠΈ μ˜€μŠ΅λ‹ˆλ‹€. 1. 고객 λ§žμΆ€ν˜• μ—¬ν–‰μΆ”μ²œ μ‚¬μ΄νŠΈ μ œμž‘ 2. λ°˜μ‘ν˜• μ›Ήμ‚¬μ΄νŠΈ κ΅¬ν˜„ 3. λ°μ΄ν„°λ² μ΄μŠ€ ν™œμš© 및 λΉ„μ •ν˜• 데이터(μ•‘μ„ΈμŠ€ 둜그) 뢄석 4. λΆ„μ‚°νŒŒμΌμ‹œμŠ€ν…œ Hadoop ν™œμš© 5. μ‹€μ‹œκ°„ 톡계 및 μΈμ‚¬μ΄νŠΈ λ„μΆœ [Method &amp; Process] λ³Έ ν”„λ‘œμ νŠΈλŠ” μ‚¬μš©μž μ€‘μ‹¬μ˜ tour ν™ˆνŽ˜μ΄μ§€λ₯Ό λ§Œλ“€μ–΄ μ—¬λŸ¬ μ†Œν”„νŠΈμ›¨μ–΄ 툴과 κΈ°λ°˜μ„ μ‘μš©ν•˜μ—¬ 섀계뢀터 μ‹€ν–‰λ‹¨κ³„κΉŒμ§€μ˜ 과정을 거쳐 ν”„λ‘œμ νŠΈλ₯Ό μ§„ν–‰ν•˜μ˜€μŠ΅λ‹ˆλ‹€. Customer-Centered Tour λΌλŠ” 주제λ₯Ό μ •ν•˜μ—¬ μ—¬ν–‰μƒν’ˆμ„ κ΄€λ¦¬μžμ—κ²Œ μ‚¬μš©μžκ°€ μ›ν•˜λŠ” 정보λ₯Ό μž…λ ₯ν•˜κ³  μ‹ μ²­ν•˜μ—¬ κ·Έ 정보(데이터)λ₯Ό κ΄€λ¦¬μžκ°€ 그에 λ§žλŠ” 여행사 νŒ¨ν‚€μ§€μƒν’ˆμ„ 제곡 ν•˜λ„λ‘ μ„€κ³„ν•˜μ˜€μŠ΅λ‹ˆλ‹€. MVCλͺ¨λΈμ„ 기점으둜 ν™˜κ²½μ„€μ • ν”„λ‘œκ·Έλž¨, μ†Œν”„νŠΈμ›¨μ–΄, 뢄석틀을 μ΄μš©ν•˜μ—¬ ν”„λ‘œμ νŠΈλ₯Ό 연ꡬ 및 μ§„ν–‰ν•˜μ˜€μœΌλ©°, Customer-Based Tour ν™ˆνŽ˜μ΄μ§€λ₯Ό μ„€κ³„ν•˜μ—¬ μ†Œν”„νŠΈμ›¨μ–΄ μ€‘μ‹¬μœΌλ‘œ JAVA와 같은 객체지ν–₯언어와 κ΄€κ³„ν˜• λ°μ΄ν„°λ² μ΄μŠ€(Oracle11g)의 ν™˜κ²½μ„€μ •μ΄λ‚˜ (Linux) μ†Œν”„νŠΈμ›¨μ–΄λ₯Ό μ—°λ™ν•˜μ—¬ λΆ„μ„ν‹€λ‘œ 데이터λ₯Ό λΆ„μ„ν•˜λŠ” ν”„λ‘œκ·Έλž˜λ°(R, ν•˜λ‘‘)을 μ΄μš©ν•˜μ—¬ μž…μΆœλ ₯ 된 데이터λ₯Ό λΆ„μ„ν•˜μ—¬ ν”„λ‘œμ νŠΈλ₯Ό μ™„λ£Œν•˜μ˜€μŠ΅λ‹ˆ. 각 ν”„λ‘œκ·Έλž¨κ³Ό μ• ν”Œλ¦¬μΌ€μ΄μ…˜νˆ΄μ„ μ†Œκ°œν•˜κ³  μ‘μš© 및 μ—°κ΅¬μ§„ν–‰μ ˆμ°¨λ₯Ό μž‘μ„±ν•˜μ˜€μŠ΅λ‹ˆλ‹€. Customer Mode μ‚¬μš©μžκ°€ 메인 νŽ˜μ΄μ§€(index)λ₯Ό 기점으둜 νšŒμ›κ°€μž… νŽ˜μ΄μ§€λ₯Ό 톡해 아이디λ₯Ό λ§Œλ“€μ–΄ νšŒμ›μΈμ¦μ„ ν•˜μ—¬ Customer mode둜 μ ‘μ†ν•©λ‹ˆλ‹€. Customer Modeμ—μ„œλŠ” ν™ˆνŽ˜μ΄μ§€ μƒμœ„μ— 더 λ§Žμ€ μΉ΄ν…Œκ³ λ¦¬λ₯Ό 찾을 수 μžˆμŠ΅λ‹ˆλ‹€. λ˜ν•œ 메인 νŽ˜μ΄μ§€ ν•˜λ‹¨μ—μ„œ μ‚¬μš©μžκ°€ μ›ν•˜λŠ” 각 λ‚˜λΌλ³„ 여행지λ₯Ό ν΄λ¦­ν•˜μ—¬ λ‹€μŒνŽ˜μ΄μ§€μΈ λ‚˜λΌλ³„μƒμ„Έμ •λ³΄λ₯Ό λ³Ό 수 있음과 λ™μ‹œμ— νŒ¨ν‚€μ§€μƒν’ˆμ„ μ‹ μ²­ν•  수 μžˆλŠ” νŽ˜μ΄μ§€λ‘œ 가도둝 링크λ₯Ό μ„€κ³„ν•˜μ˜€μŠ΅λ‹ˆλ‹€. λ˜ν•œ μ‚¬μš©μžλŠ” ν™ˆνŽ˜μ΄μ§€μ—μ„œ μ œκ³΅ν•˜λŠ” 여행지 톡계λ₯Ό 각 λ‚˜λΌλ³„, ν…Œλ§ˆλ³„, 계쑀별, μ—°λ Ήλ³„λ‘œ λ³Ό 수 μžˆλ„λ‘ μ œκ³΅λ©λ‹ˆλ‹€. Administrator Mode νŒ¨ν‚€μ§€μƒν’ˆμ„ λ“±λ‘ν•˜λ©΄ κ·Έ 정보가 κ΄€λ¦¬μž(Administrator)μ—κ²Œ λ‘œκ·Έλ°μ΄ν„°κ°€ μž…λ ₯되고 여행사별 νŒ¨ν‚€μ§€μƒν’ˆκ³Ό μ‚¬μš©μž(Customer Mode)μ—κ²Œ 연결을 ν•˜μ—¬ λ‹€μ‹œ μ‚¬μš©μžμ—κ²Œ λ³΄λ‚΄μ€λ‹ˆλ‹€. κ΄€λ¦¬μžλŠ” VIP고객정보, νšŒμ›κ΄€λ¦¬ 등을 μœ„ν•œ νŽ˜μ΄μ§€μ™€ Rκ³Ό ν•˜λ‘‘μ„ 톡해 λΆ„μ„ν•œ ν†΅κ³„μžλ£Œλ₯Ό μœ„ν•œ νŽ˜μ΄μ§€ 등을 확인 ν•  수 μžˆμŠ΅λ‹ˆλ‹€. [Results &amp; discussions] [ν•œκ³„] μ›Ήμ–΄ν”Œλ¦¬μΌ€μ΄μ…˜ κ΄€λ ¨ ν•˜μ—¬μ„œλŠ” μ‚¬μš©μž λ™μ‹œ 접속은 둜그인 μƒνƒœλ₯Ό μœ μ§€ν•˜κ³  μžˆλŠ” μ„Έμ…˜(session)의 갯수λ₯Ό μΈ‘μ •ν•˜λŠ” κ²ƒμž…λ‹ˆλ‹€. ν•˜μ§€λ§Œ μ›Ήμ„œλΉ„μŠ€λŠ” μ‚¬μš©μžκ°€ λ‘œκ·Έμ•„μ›ƒν•œ μ‹œμ μ„ νŒŒμ•…ν•˜κΈ°λŠ” μ–΄λ €μ› μœΌλ©° μ„œλ²„μ˜ 지원 상황(λ©”λͺ¨λ¦¬,CPU,λ„€νŠΈμ›Œν¬)에 따라 μ„±λŠ₯ 편차(였차)κ°€ λ°œμƒν•˜μ˜€μŠ΅λ‹ˆλ‹€.μˆœκ°„ μ²˜λ¦¬λŸ‰μœΌλ‘œ μ„œλ²„μ˜ λͺ¨λ“  μ„±λŠ₯을 평가할 수 μ—†λ‹€λŠ” 결둠이 λ‚˜μ™”μŠ΅λ‹ˆλ‹€. μ„±λŠ₯평가 μ‚¬μš©μž μž…μž₯μ—μ„œλŠ” μ„œλ²„ μ‘λ‹΅μ‹œκ°„μ΄ μ§§μ„μˆ˜λ‘ μ’‹μŒ μ„œλ²„κ°€ 아무리 λΉ λ₯΄λ”라도 μ„œλ²„μ™€ μ‚¬μš©μž μ‚¬μ΄μ—λŠ” λ„€νŠΈμ›Œν¬ νšŒμ„ μ΄ μ‘΄μž¬ν•˜κΈ° λ•Œλ¬Έμ— 지연 μ‹œκ°„(latency time)이 λ°œμƒν•  수 밖에 μ—†μŒ. μ„œλ²„ 섀계 및 κ΄€λ¦¬μž μž…μž₯μ—μ„œλŠ” μ„œλ²„ 쀑단(halt)κ°€ κ°€μž₯ 큰 λ¬Έμ œκ°€ 될 수 μžˆλ‹€. μ‚¬μš©μžμ˜ μ΅œμ ν•œ μ„œλΉ„μŠ€ κ²½ν—˜μ„ μœ„ν•œ &quot;짧은 응닡 μ‹œκ°„(short response time)&quot; κ³Ό μ„œλ²„μ˜ μ•ˆμ •μ μΈ μš΄μ˜μ„ μœ„ν•œ &quot;μ μ ˆν•œ μ²˜λ¦¬λŸ‰(proper throughtput)&quot;이 μ„±λŠ₯μ§€ν‘œμ˜ κ°€μž₯ 큰 기쀀이 λœλ‹€κ³  ν•  수 μžˆλ‹€. [μ—­λŸ‰] λ³Έ ν”„λ‘œμ νŠΈλ‘œλΆ€ν„° 빅데이터λ₯Ό μ΄μš©ν•œ λ‹€μ–‘ν•œ 뢄석 μ—­λŸ‰ 및 μ‹œμŠ€ν…œ κ΅¬μΆ•μ—­λŸ‰μ„ μŠ΅λ“ν•œ ν›„ κΈ°μ‘΄μ‹œμŠ€ν…œμ—μ„œ λ°œμƒν•œ λ‹€λŸ‰μ˜ 데이터λ₯Ό ν™œμš©ν•˜μ—¬ μ—¬λŸ¬ λΆ„μ•Όμ˜ λΉ„μ¦ˆλ‹ˆμŠ€κ°€ 진행될 수 μžˆμŒμ„ μ•Œκ²Œλ˜μ—ˆμŠ΅λ‹ˆλ‹€. λ¦¬λˆ…μŠ€ OS/κΈ°λ³Έμ§€μ‹μŠ΅λ“μœΌλ‘œλŠ” ν˜„μ—…μ—μ„œ μ• μš©ν•˜λŠ” λ¦¬λˆ…μŠ€λ₯Ό κ°œλ³„ PC에 μ„€μΉ˜/ν™œμš©ν•˜μ—¬ λŒ€ν˜•μ‹œμŠ€ν…œ 운영λŠ₯λ ₯을 μŠ΅λ“ν•˜κ³  λ¦¬λˆ…μŠ€ OS기반의 ν”„λ‘œκ·Έλž¨ 개발 λŠ₯λ ₯ 및 SQL μ‚¬μš© λŠ₯λ ₯을 μŠ΅λ“ν•˜μ˜€μŠ΅λ‹ˆλ‹€. 빅데이터 μ‹œμŠ€ν…œκ³Ό μ‘μš© μ• ν”Œλ¦¬μΌ€μ΄μ…˜ μ—°λ™μœΌλ‘œλŠ” 빅데이터 ν™˜κ²½μ—μ„œ λΆ„μ„λœ 자료λ₯Ό WebApplication 및 λ‹€μ–‘ν•œ Application 과의 연동을 톡해 뢄석 자료 Visualization κΈ°μˆ μ„ μŠ΅λ“ν•˜μ˜€μŠ΅λ‹ˆλ‹€. 빅데이터 μ‹œμŠ€ν…œ ꡬ좕/ λ‹€μ–‘ν•œ 빅데이터 뢄석 μ—­λŸ‰ μŠ΅λ“ν•˜μ˜€μŠ΅λ‹ˆλ‹€. 빅데이터 ν™˜κ²½μ„ μœ„ν•œ ν•˜λ‘‘ μ‹œμŠ€ν…œ ꡬ좕과 그에 μ—°κ΄€λœ ν•˜λ‘‘μ—μ½” μ‹œμŠ€ν…œμ„ ꡬ좕 ν•˜κ³  λ‹€μ–‘ν•œ ν˜„μ—…μ—μ„œ ν•„μš”λ‘œ ν•˜λŠ” 뢄석전문가가 되기 μœ„ν•΄ 빅데이터 뢄석 기법/μ „λž΅ μ‹€μŠ΅μ„ μ§„ν–‰ν•˜μ˜€μŠ΅λ‹ˆλ‹€. [Conclusion] λ³Έ ν”„λ‘œμ νŠΈμ™€ 같이 ν™ˆνŽ˜μ΄μ§€λ₯Ό κ°œμ„€, 운영 및 μœ μ§€ν•˜λŠ” κ³Όμ •κ³Ό λ”λΆˆμ–΄ μ‹€μ‹œλ‘œ λΆ„μ„ν•˜λŠ” 과정을 ν†΅ν•˜μ—¬ 톡합적이고 λŠκΉ€ μ—†λŠ” κ΄€λ¦¬μž-μ†ŒλΉ„μž 체인의 λΈŒλžœλ“œ 관계 ν˜•μ„±μ„ 보닀 μš©μ΄ν•˜κ²Œ ν•  수 μžˆλ‹€. μ™œλƒν•˜λ©΄ 운영자(κ΄€λ¦¬μž)λŠ” μ‹€μ‹œκ°„ λ°μ΄ν„°λ‘œκ·Έλ₯Ό ν†΅ν•˜μ—¬ μ†ŒλΉ„μžμ˜ ν™ˆνŽ˜μ΄μ§€μ—μ„œ ν–‰ν•˜λŠ” ν–‰μœ„λ“€μ„ λ°μ΄ν„°λ‘œ μ•Œ 수 있기 λ•Œλ¬Έμ΄λ‹€. μ—¬ν–‰μƒν’ˆμ΄ ꡬ글과 같은 ν¬ν„Έμ‚¬μ΄νŠΈμ˜ μ•Œκ³ λ¦¬μ¦˜μ— 잘 맞좰져 검색이 잘 λ˜λ„λ‘ λ…Έλ ₯ν•΄μ•Ό ν•˜λ©° μ½˜ν…μΈ  확산성을 μœ„ν•œ 연ꡬ 및 개발이 ν•„μš”ν•˜λ‹€. μ†ŒλΉ„μžλ“€μ€ 정보λ₯Ό κ²€μƒ‰ν•˜κ³ , κ΄€λ¦¬μžλŠ” μ†ŒλΉ„μžμ—κ²Œ 정보λ₯Ό μ œκ³΅ν•˜λŠ” λ°©μ‹μœΌλ‘œμ„œ 이 λ‘˜μ„ λ§€ν•‘ν•˜λŠ” λ…Έλ ₯이 ν•„μš”ν•˜κ²Œ 될 것이며 λΉ„μ •ν˜•ν™”λœ 데이터λ₯Ό μ΄μš©κ°€λŠ₯ν•œ μ •ν˜•λ°μ΄ν„°λ‘œ λ§Œλ“œλŠ” μž‘μ—…μ„ μš©μ΄ν•˜κ²Œ ν•˜μ—¬ λ”œλ ˆλ§ˆλ₯Ό μ΅œμ†Œν™”ν•˜μ—¬ ν•œλ‹€.
[kor ver.]Global GO (Bigdata-Cloud computing project - mainly in MVC model2) from Adonis Han
]]> 107 1 https://cdn.slidesharecdn.com/ss_thumbnails/random-171227155714-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 House pricing prediction in R(Regression Project) /slideshow/house-pricing-prediction-in-rregression-project-85099508/85099508 reportprojectrregressionkingcountyrfairy-171227155123
Regression project in R data : Kingcounty ( Kaggle data) Process 1. EDA -stepwise 2. analyse -correlation -t-test -F-test 3. modeling -R squared 4. validation -VIF -Multicollinearity ]]>
Regression project in R data : Kingcounty ( Kaggle data) Process 1. EDA -stepwise 2. analyse -correlation -t-test -F-test 3. modeling -R squared 4. validation -VIF -Multicollinearity ]]> Wed, 27 Dec 2017 15:51:23 GMT /slideshow/house-pricing-prediction-in-rregression-project-85099508/85099508 ssuserf03d2b@slideshare.net(ssuserf03d2b) House pricing prediction in R(Regression Project) ssuserf03d2b Regression project in R data : Kingcounty ( Kaggle data) Process 1. EDA -stepwise 2. analyse -correlation -t-test -F-test 3. modeling -R squared 4. validation -VIF -Multicollinearity <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/reportprojectrregressionkingcountyrfairy-171227155123-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> Regression project in R data : Kingcounty ( Kaggle data) Process 1. EDA -stepwise 2. analyse -correlation -t-test -F-test 3. modeling -R squared 4. validation -VIF -Multicollinearity
House pricing prediction in R(Regression Project) from Adonis Han
]]> 297 3 https://cdn.slidesharecdn.com/ss_thumbnails/reportprojectrregressionkingcountyrfairy-171227155123-thumbnail.jpg?width=120&height=120&fit=bounds document Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Facial detection by CNN(Convolution Neural Network) in Kaggle /slideshow/facial-detection-by-cnnconvolution-neural-network-in-kaggle/82565241 facialdetectionado-171123070427
Facial detection by CNN(Convolution Neural Network) in Kaggle dataset : kaggle(https://www.kaggle.com/c/facial-keypoints-detection/) github: https://github.com/AdonisHan/Facial-dectection-Pyfairy Package: Lasagna, Keras, TensorFlow, OpenCV Model Class: SVM, Neural Network, Convolutional NN RMSE score private score : 1.645 public score : 1.872]]>
Facial detection by CNN(Convolution Neural Network) in Kaggle dataset : kaggle(https://www.kaggle.com/c/facial-keypoints-detection/) github: https://github.com/AdonisHan/Facial-dectection-Pyfairy Package: Lasagna, Keras, TensorFlow, OpenCV Model Class: SVM, Neural Network, Convolutional NN RMSE score private score : 1.645 public score : 1.872]]> Thu, 23 Nov 2017 07:04:27 GMT /slideshow/facial-detection-by-cnnconvolution-neural-network-in-kaggle/82565241 ssuserf03d2b@slideshare.net(ssuserf03d2b) Facial detection by CNN(Convolution Neural Network) in Kaggle ssuserf03d2b Facial detection by CNN(Convolution Neural Network) in Kaggle dataset : kaggle(https://www.kaggle.com/c/facial-keypoints-detection/) github: https://github.com/AdonisHan/Facial-dectection-Pyfairy <Method> Package: Lasagna, Keras, TensorFlow, OpenCV Model Class: SVM, Neural Network, Convolutional NN <Final Model> RMSE score private score : 1.645 public score : 1.872 <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/facialdetectionado-171123070427-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> Facial detection by CNN(Convolution Neural Network) in Kaggle dataset : kaggle(https://www.kaggle.com/c/facial-keypoints-detection/) github: https://github.com/AdonisHan/Facial-dectection-Pyfairy Package: Lasagna, Keras, TensorFlow, OpenCV Model Class: SVM, Neural Network, Convolutional NN RMSE score private score : 1.645 public score : 1.872
Facial detection by CNN(Convolution Neural Network) in Kaggle from Adonis Han
]]> 743 2 https://cdn.slidesharecdn.com/ss_thumbnails/facialdetectionado-171123070427-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 https://cdn.slidesharecdn.com/profile-photo-ssuserf03d2b-48x48.jpg?cb=1658282340 Hello I take advantage of a great opportunity to share ideas in here! I could build partnering through friendly, attentive service to YOU. naver.blog.com/sanghan1990 https://cdn.slidesharecdn.com/ss_thumbnails/fairieslda-181022042645-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/lda-latent-dirichlet-allocation-fairies-nlp-series-korean-ver/120255090 LDA : latent Dirichlet... https://cdn.slidesharecdn.com/ss_thumbnails/nlptutorialfairies-180309032040-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/kor-vernlp-embeddingword2vec-tutorial-implementationtensorflow/90102946 (Kor ver.)NLP embeddin... https://cdn.slidesharecdn.com/ss_thumbnails/fairies-wavenet-180309031321-thumbnail.jpg?width=320&height=320&fit=bounds ssuserf03d2b/how-to-understand-and-implement-the-wavenet how to understand and ...