Page 15 -

1.3.1 ా҅೟җ ݠन ۞׬

ా҅੄ ৉ࢎח ӝਗ੹ 5ࣁӝ ୡө૑ Ѣठ۞ ৢۄо૑݅ 19ࣁӝ ݈ _ 20ࣁӝ ୡী Ѧ୛ ١੢ೠ

ా҅೟੗ ࣁ ݺ ೐ےदझ Ҏఢ ண ೖযट ۽օ٘ ೖ࣊ ী ੄೧ ખ؊ ୓҅੸ੋ ೟ޙਵ۽ ੿ܻغ
঻׮ ੉ٜ਷ օܻ ঌ۰૓ ా҅ ӝߨੋ ഥӈ ࠙ࢳ(regression) ਋ب(likelihood) ࠙࢑ ࠙ࢳ(analysis of
variance) ࢚ҙ ࠙ࢳ(correlation) ١ਸ Ҋউ೮׮

20ࣁӝ ઺߈ী ੉ܰ۞ ా҅೟਷ ؘ੉ఠ ࠙ࢳҗ ӟ޻ೞѱ ৊੉ӝ द੘೮׮ 1962֙ী ߊ಴ػ ਬݺ
1
ೠ ֤ޙ ؘ੉ఠ ࠙ࢳ੄ ޷ې(The Future of Data Analysis) ীࢲ ੷੗ ઓ ఠః ޷Ҵ੄ ࣻ೟੗੉੗ ా
҅೟੗੉ݴ Ҋࣘ ಹܻী ߸ജ ߅झ ର౟ HSD Ѩ੿੄ ߊݺ੗۽ ੜ ঌ۰ઉ ੓׮ ח tѾҴ ա੄

о੢ ௾ ҙब਷ ؘ੉ఠ ࠙ࢳী ੓׮uҊ ঱ә೮ਵݴ ੉ח ਽ਊ ా҅೟੄ ઺ਃೠ ੉੿಴۽ ߉ইٜ
ৈ ઎׮

Ӓ റ۽ ࣻभ ֙р ా҅೟੗ٜ਷ ਽ਊ ҅࢑ ా҅೟ ࠙ঠী ݆਷ ҙबਸ ࠁ੉ݴ ҙ۲ োҳܳ ૑ࣘ
೮׮ ೞ૑݅ Ӓ ׼द ా҅೟੗ٜ੄ ֢۱਷ ஹೊఠ җ೟ ࠙ঠ੄ ݠन ۞׬ োҳ৬ ࢚׼൤ زڄয
ઉ ੓঻׮

1950֙؀ റ߈ ஹೊఠ੄ ߊ੹੉ ਬইӝী ੽যٜݶࢲ ஹೊఠ җ೟੗ٜ਷ ࢎۈ੄ न҃ ੹׳ ݽ؛
ীࢲ ଱উೠ ੋҕ૑מ दझమ ૊ ੋҕ न҃ݎ(neural networks)ਸ োҳೞӝ द੘೮׮ ೐ۘ௼ ۽ઃ࠶
ۖ੄ ࢶҳ੸ੋ ಌࣆ౟ۿ(perceptron) ঌҊ્ܻҗ Ӓ ٍܳ ੉਷ ਤ٘۽৬ ഐ೐੄ োҳח ࢜۽਍ ࠙ঠ

ী ؀ೠ োҳ੗ٜ੄ ಩ߊ੸ੋ ҙबਸ ࠛ۞ੌਵௌ׮ 2

ੋҕ न҃ݎ੄ ୡӝ ࢿҕী ٍ੉য ࣻभ ֙р ؘ੉ఠ۽ࠗఠ ಁఢਸ ੗زਵ۽ ೟णೞח ׮নೠ
ݠन ۞׬ ӝߨ ୭Ӕ੽ ੉ਓ(nearest neighbor) ੄ࢎ Ѿ੿ ౟ܻ(decision tree) k-ಣӐ ҵ૘ച(k-means
clustering) ࢲನ౟ ߭ఠ ݠन(Support Vector Machine, SVM) ١ ੉ Ҋউغ঻׮

ஹೊఠ दझమ੉ ؊ ࡈۄ૑Ҋ ࠺ਊ੉ ੷۴೧૑ݶࢲ ؊ ݆਷ ؘ੉ఠࣇਸ ݠन ۞׬ ӝߨী ഝਊೡ
ࣻ ੓ѱ غ঻׮ ੉۞ೠ ߊ੹਷ ؊ ѼҊೠ ݠन ۞׬ ঌҊ્ܻ੄ ҳഅਵ۽ ੉য઎׮

1989֙ GTE োҳࣗ੄ ؘ੉ఠ җ೟੗৓؍ ӒۨҊܻ ೖইపझః-ࢥೖ۽ח KDD(Knowledge
Discovery in Database) ਕ௼ࣸਸ ѐ୭೮׮ KDD ਕ௼ࣸ਷ Әࣁ ਬݺ೧ઉ ݒ֙ KDD Cup ؘ੉ఠ
݃੉׬ ؀ഥܳ ઱ҙೞח ACM-SIGKDD ೟ഥ۽ ߊ੹೮׮

1 IUUQ XXX TUBOGPSE FEV _HBWJTI EPDVNFOUT 5VLFZ@UIF@GVUVSF@PG@EBUB@BOBMZTJT QEG
2 झగನ٘ Үࣻ৓؍ ߡա٘ ਤ٘۽৬ Ӓ੄ ઁ੗ ప٘ ഐ೐ח "%"-*/&੉ۄח न҃ݎ ݽ؛ਸ ઁউ೮׮

026

하둡과 스파크를 활용한 실용 데이터과학(본문)5차.indd 26 2017-07-25 오후 4:15:43