Page 17 -

ҳӖҗ ঠറ അ੤ח ࡂ ח Ѩ࢝ ূ૓ Ѩ࢝য ઁউ Ү੿ ӝࣿ੄ ࢿמਸ ೱ࢚ೞח ؘ ؀ӏݽ
ؘ੉ఠ ӝ߈ ঌҊ્ܻਸ ࢎਊೠ׮
Ѣ؀ ੋఠ֔ Ѩ࢝ ഥࢎח ಕ੉૑ ࠭৬ ௿ܼ ੿ࠁܳ ࠙ࢳ೧ Ѩ࢝ ূ૓ ࢎਊ੗੄ ௿ܼܫਸ ৘
ஏೞҊ ࢎਊ੗৬ ҙ۲ࢿ੉ ֫਷ ৡۄੋ ҟҊܳ ֢୹ೠ׮

݂௼٘ੋҗ ಕ੉झ࠘਷ ࢎਊ੗ р ࢎഥ੸ ҙ҅ Ӓې೐ܳ ࠙ࢳ೧ r׼न੉ ঌ ࣻب ੓ח ࢎۈ
(People You May Know, PYMK)s੉ۄח ஘ҳ ୶ୌ ӝמਸ ઁҕೠ׮

֔೒ܼझ ੉߬੉ ই݃ઓ਷ ҟߧਤೠ ؘ੉ఠܳ ӝ߈ਵ۽ ؊ ա਷ ࢎਊ੗ ҃೷ਸ ઁҕೞ۰
Ҋ ੗زചػ ઁಿ ژח ৔ച ୶ୌ ӝמਸ ઁҕೠ׮

ಕ੉౹਷ ૑ࠛ ࢎӝܳ ఐ૑ೞ۰Ҋ ؀ӏݽ Ӓې೐ ঌҊ્ܻਸ ੸ਊೠ׮

׼द ੋఠ֔ Ѣੋٜ਷ ࢶҳ੸ੋ ࠺੹ਸ ઁदೞݴ Ӓٜ੉ ࠁਬೠ ؀ӏݽ ਗद ؘ੉ఠࣇ(raw dataset)

ਸ ؊਌ ഄन੸੉Ҋ ࢜۽਍ ߑߨਵ۽ ഝਊೡ ࣻ ੓ח оמࢿਸ ߊҷ೮׮ ژೠ ੋఠ֔ীࢲ ߊࢤೞ
ח ؀ӏݽ ؘ੉ఠܳ ୭؀ೠ ഝਊೞ۰ݶ ݆ࣻ਷ ӝࣿ੸ য۰਑ਸ ӓࠂ೧ঠ ೠ׮ח ࢎपਸ ־ҳࠁ
׮ ࡈܻ ӵ׳ও׮ Ӓ Ѿҗ ੋఠ֔ ӝসٜ਷ ৈ۞ ࢜۽਍ ӝࣿҗ ࣗ೐౟ਝয بҳ ҳӖ ౵ੌ दझ
మ(GFS) ݗܻٓझ(MapReduce) ೞن(Hadoop) ೖӒ(Pig) ೞ੉࠳(Hive) ஠࢑٘ۄ(Cassandra) झ౵௼

(Spark) झ఻(Storm) HBase ١ ܳ ѐߊ೧ ࠻ؘ੉ఠ ࠙ঠী ഄन੄ ౵بܳ ੌਵௌ׮

1.3.3 അ؀ ӝস੄ ؘ੉ఠ җ೟

ੋఠ֔ Ѣੋٜ੉ оઉৡ ഄन ؋࠙ী ࢚ਊ بҳҗ য়೑ ࣗझ ઁಿ ݽفীࢲ ৈ۞ ೨ब ӝࣿਸ ࢎ
ਊೞѱ غ঻׮

୐૩ ӝস਷ ؀ӏݽ ؘ੉ఠܳ ࣻ૘ೞҊ ੷੢ೞӝ ਤೠ ઁ߈ ࠺ਊ੉ ੺хغ঻׮ ੷۴ೞҊ ࡅܲ
झషܻ૑ ӝࣿ ௿۞झఠ ஹೊ౴ ӝࣿ ೞن э਷ য়೑ ࣗझ ࣗ೐౟ਝয੄ ߊ੹੉ ઱بೠ Ѿҗۄ
Ҋ ೡ ࣻ ੓׮ ؘ੉ఠח ੉ઁ ӝস੄ о஖ ੓ח ੗࢑੉Ҋ ؘ੉ఠ ೙ఠ݂੉ա ࠁઓ ӝೠ੄ ઁೠ э

਷ ׮নೠ ࠺ਊ ੺х ֢۱ਸ ٜ੉૑ ঋҊب ӝস੄ ݽٚ ؘ੉ఠܳ ਗद ഋక۽ ੷੢ೡ ࣻ ੓׮ ؀
ਊ۝ ؘ੉ఠ ੷੢ ӝࣿ ؋࠙ী അ؀ ӝস਷ ੉੹ө૑ ࠛоמ೮؍ ࢜۽਍ ূఠ೐ۄ੉ૉ ؘ੉ఠ җ
೟ গ೒ܻா੉࣌ਸ ഝਊೡ ࣻ ੓׮

ل૩ R ౵੉ॆ scikit-learn झ౵௼ MLlib э਷ ؘ੉ఠ җ೟ য়೑ ࣗझ ಁః૑о ҕѐغݶࢲ ૑
Әө૑ ӝসীࢲ ഝਊೡ ࣻ হ঻؍ ݠन ۞׬җ ా҅೟ ӝ߈੄ Ҋә ؘ੉ఠ ݃੉׬ ঌҊ્ܻਸ ӝ
সٜ੄ ؀ӏݽ ؘ੉ఠࣇী औҊ ਬোೠ ߑधਵ۽ ੸ਊೠ׮ ੉۞ೠ ߸ച ؋࠙ী ӝস਷ ؘ੉ఠ

028

하둡과 스파크를 활용한 실용 데이터과학(본문)5차.indd 28 2017-07-25 오후 4:15:43