T\u1ea1i sao c\u1ea7n s\u1eed d\u1ee5ng Khoa h\u1ecdc D\u1eef li\u1ec7u?<\/strong><\/p>\n\n\n\n Nh\u00e0 khoa h\u1ecdc d\u1eef li\u1ec7u (data scientist) l\u00e0 ai?<\/strong><\/p>\n\n\n\n Nh\u00e0 khoa h\u1ecdc d\u1eef li\u1ec7u l\u00e0m g\u00ec?<\/strong><\/p>\n\n\n\n Data Science v\u00e0 Business Intelligence (BI) kh\u00e1c nhau nh\u01b0 th\u1ebf n\u00e0o?<\/strong><\/p>\n\n\n\n C\u00e1c c\u00f4ng vi\u1ec7c c\u1ee5 th\u1ec3 trong Khoa h\u1ecdc d\u1eef li\u1ec7u<\/strong><\/p>\n\n\n\n Khoa h\u1ecdc d\u1eef li\u1ec7u <\/strong>(data science) l\u00e0 g\u00ec?<\/strong><\/p>\n\n\n\n Khoa h\u1ecdc d\u1eef li\u1ec7u l\u00e0 s\u1ef1 k\u1ebft h\u1ee3p c\u1ee7a nhi\u1ec1u c\u00f4ng c\u1ee5, thu\u1eadt to\u00e1n v\u00e0 nguy\u00ean t\u1eafc h\u1ecdc m\u00e1y v\u1edbi m\u1ee5c ti\u00eau kh\u00e1m ph\u00e1 c\u00e1c m\u1eabu \u1ea9n t\u1eeb d\u1eef li\u1ec7u th\u00f4. Nh\u01b0ng \u0111i\u1ec1u n\u00e0y kh\u00e1c g\u00ec v\u1edbi c\u00f4ng vi\u1ec7c c\u00e1c nh\u00e0 th\u1ed1ng k\u00ea\/ nh\u00e0 ph\u00e2n t\u00edch d\u1eef li\u1ec7u \u0111\u00e3 l\u00e0m trong nhi\u1ec1u n\u0103m?<\/p>\n\n\n\n C\u00e2u tr\u1ea3 l\u1eddi n\u1eb1m \u1edf s\u1ef1 kh\u00e1c bi\u1ec7t gi\u1eefa gi\u1ea3i th\u00edch v\u00e0 d\u1ef1 \u0111o\u00e1n.<\/p>\n\n\n\n Nh\u00e0 th\u1ed1ng k\u00ea\/ nh\u00e0 ph\u00e2n t\u00edch d\u1eef li\u1ec7u (statistician\/ data analyst) th\u01b0\u1eddng gi\u1ea3i th\u00edch nh\u1eefng g\u00ec \u0111ang x\u1ea3y ra b\u1eb1ng c\u00e1ch ph\u00e2n t\u00edch d\u1eef li\u1ec7u l\u1ecbch s\u1eed. Trong khi \u0111\u00f3, nh\u00e0 khoa h\u1ecdc d\u1eef li\u1ec7u (data scientist) kh\u00f4ng ch\u1ec9 th\u1ef1c hi\u1ec7n ph\u00e2n t\u00edch d\u1eef li\u1ec7u \u0111\u1ec3 khai th\u00e1c th\u00f4ng tin chi ti\u1ebft t\u1eeb \u0111\u00f3, hay c\u00f2n g\u1ecdi l\u00e0 ph\u00e2n t\u00edch m\u00f4 t\u1ea3 (descriptive analytics) m\u00e0 c\u00f2n s\u1eed d\u1ee5ng c\u00e1c thu\u1eadt to\u00e1n h\u1ecdc m\u00e1y n\u00e2ng cao kh\u00e1c nhau \u0111\u1ec3 x\u00e1c \u0111\u1ecbnh kh\u1ea3 n\u0103ng xu\u1ea5t hi\u1ec7n c\u1ee7a m\u1ed9t s\u1ef1 ki\u1ec7n c\u1ee5 th\u1ec3 trong t\u01b0\u01a1ng lai. M\u1ed9t nh\u00e0 khoa h\u1ecdc d\u1eef li\u1ec7u s\u1ebd xem x\u00e9t d\u1eef li\u1ec7u t\u1eeb nhi\u1ec1u g\u00f3c \u0111\u1ed9, \u0111\u00f4i khi l\u00e0 nh\u1eefng g\u00f3c \u0111\u1ed9 kh\u00f4ng \u0111\u01b0\u1ee3c bi\u1ebft \u0111\u1ebfn t\u1eeb tr\u01b0\u1edbc.<\/p>\n\n\n\n N\u00f3i c\u00e1ch kh\u00e1c, Khoa h\u1ecdc d\u1eef li\u1ec7u ch\u1ee7 y\u1ebfu \u0111\u01b0\u1ee3c s\u1eed d\u1ee5ng \u0111\u1ec3 d\u1ef1 \u0111o\u00e1n v\u00e0 h\u1ed7 tr\u1ee3 vi\u1ec7c \u0111\u01b0a ra quy\u1ebft \u0111\u1ecbnh b\u1eb1ng c\u00e1ch s\u1eed d\u1ee5ng ph\u00e2n t\u00edch d\u1ef1 \u0111o\u00e1n nguy\u00ean nh\u00e2n – k\u1ebft qu\u1ea3 (predictive causal analytics), ph\u00e2n t\u00edch \u0111\u1ec1 xu\u1ea5t (prescriptive analytics – d\u1ef1 \u0111o\u00e1n c\u1ed9ng v\u1edbi \u0111\u1ec1 xu\u1ea5t h\u00e0nh \u0111\u1ed9ng ph\u00f9 h\u1ee3p) v\u00e0 h\u1ecdc m\u00e1y (machine learning).<\/p>\n\n\n\n Ph\u00e2n t\u00edch <\/strong>d\u1ef1 \u0111o\u00e1n nguy\u00ean nh\u00e2n<\/strong> – k\u1ebft qu\u1ea3<\/strong> (<\/strong>predictive causal analytics<\/strong>)<\/strong> \u2013 l\u00e0 m\u00f4 h\u00ecnh s\u1eed d\u1ee5ng \u0111\u1ec3 d\u1ef1 \u0111o\u00e1n c\u00e1c kh\u1ea3 n\u0103ng c\u1ee7a m\u1ed9t s\u1ef1 ki\u1ec7n c\u1ee5 th\u1ec3 trong t\u01b0\u01a1ng lai. V\u00ed d\u1ee5, n\u1ebfu b\u1ea1n l\u00e0 ng\u01b0\u1eddi cho vay t\u00edn d\u1ee5ng, b\u1ea1n s\u1ebd quan t\u00e2m \u0111\u1ebfn kh\u1ea3 n\u0103ng kh\u00e1ch h\u00e0ng th\u1ef1c hi\u1ec7n c\u00e1c kho\u1ea3n thanh to\u00e1n \u0111\u00fang h\u1ea1n traong t\u01b0\u01a1ng lai. Nh\u01b0 v\u1eady, b\u1ea1n c\u00f3 th\u1ec3 s\u1ebd c\u1ea7n x\u00e2y d\u1ef1ng m\u1ed9t m\u00f4 h\u00ecnh ph\u00e2n t\u00edch d\u1ef1 \u0111o\u00e1n v\u1ec1 l\u1ecbch s\u1eed thanh to\u00e1n c\u1ee7a kh\u00e1ch h\u00e0ng \u0111\u1ec3 d\u1ef1 \u0111o\u00e1n li\u1ec7u trong t\u01b0\u01a1ng lai h\u1ecd c\u00f3 thanh to\u00e1n \u0111\u00fang h\u1ea1n hay kh\u00f4ng.<\/p>\n\n\n\n Ph\u00e2n t\u00edch \u0111\u1ec1 xu\u1ea5t (prescriptive analytics<\/strong>):<\/strong> l\u00e0 m\u1ed9t m\u00f4 h\u00ecnh c\u00f3 kh\u1ea3 n\u0103ng t\u1ef1 quy\u1ebft \u0111\u1ecbnh th\u00f4ng minh v\u00e0 kh\u1ea3 n\u0103ng s\u1eeda \u0111\u1ed5i m\u00f4 h\u00ecnh \u0111\u00f3 v\u1edbi c\u00e1c tham s\u1ed1 \u0111\u1ed9ng. N\u00f3i m\u1ed9t c\u00e1ch kh\u00e1i qu\u00e1t, ph\u01b0\u01a1ng ph\u00e1p n\u00e0y kh\u00f4ng ch\u1ec9 \u0111\u01b0a ra d\u1ef1 \u0111o\u00e1n m\u00e0 c\u00f2n \u0111\u1ec1 xu\u1ea5t m\u1ed9t lo\u1ea1t c\u00e1c h\u00e0nh \u0111\u1ed9ng \u0111\u01b0\u1ee3c x\u00e1c l\u1eadp s\u1eb5n v\u1edbi m\u1ed7i d\u1ef1 \u0111o\u00e1n v\u00e0 c\u00e1c k\u1ebft qu\u1ea3 li\u00ean quan.<\/p>\n\n\n\n M\u1ed9t trong nh\u1eefng \u1ee9ng d\u1ee5ng c\u1ee7a ph\u00e2n t\u00edch \u0111\u1ec1 xu\u1ea5t l\u00e0 tr\u00ean xe \u00f4 t\u00f4 t\u1ef1 l\u00e1i. M\u00e1y t\u00ednh tr\u00ean xe t\u1ef1 l\u00e1i s\u1ebd \u0111\u01b0\u1ee3c \u201cd\u1ea1y\u201d b\u1eb1ng d\u1eef li\u1ec7u thu th\u1eadp c\u1ee7a c\u00e1c ph\u01b0\u01a1ng ti\u1ec7n kh\u00e1c nhau. C\u00e1c thu\u1eadt to\u00e1n s\u1ebd \u0111\u01b0\u1ee3c thi\u1ebft l\u1eadp tr\u00ean c\u00e1c d\u1eef li\u1ec7u n\u00e0y. \u0110i\u1ec1u n\u00e0y s\u1ebd cho ph\u00e9p xe c\u1ee7a b\u1ea1n \u0111\u01b0a ra c\u00e1c quy\u1ebft \u0111\u1ecbnh nh\u01b0 khi n\u00e0o n\u00ean r\u1ebd, \u0111i \u0111\u01b0\u1eddng n\u00e0o, khi n\u00e0o n\u00ean gi\u1ea3m t\u1ed1c \u0111\u1ed9 ho\u1eb7c t\u0103ng t\u1ed1c \u0111\u1ed9.<\/p>\n\n\n\n H\u1ecdc m\u00e1y \u0111\u1ec3 \u0111\u01b0a ra d\u1ef1 \u0111o\u00e1n<\/strong> (machine learning for making predictions)<\/strong> – N\u1ebfu b\u1ea1n c\u00f3 d\u1eef li\u1ec7u giao d\u1ecbch c\u1ee7a m\u1ed9t c\u00f4ng ty t\u00e0i ch\u00ednh v\u00e0 c\u1ea7n x\u00e2y d\u1ef1ng m\u00f4 h\u00ecnh \u0111\u1ec3 x\u00e1c \u0111\u1ecbnh xu h\u01b0\u1edbng trong t\u01b0\u01a1ng lai, th\u00ec thu\u1eadt to\u00e1n h\u1ecdc m\u00e1y l\u00e0 l\u1ef1a ch\u1ecdn t\u1ed1t nh\u1ea5t. Tr\u01b0\u1eddng h\u1ee3p n\u00e0y g\u1ecdi l\u00e0 h\u1ecdc m\u00e1y c\u00f3 gi\u00e1m s\u00e1t – b\u1ea1n \u0111\u00e3 c\u00f3 s\u1eb5n d\u1eef li\u1ec7u \u0111\u1ec3 d\u1ef1a v\u00e0o \u0111\u00f3 \u0111\u00e0o t\u1ea1o m\u00e1y c\u1ee7a m\u00ecnh. V\u00ed d\u1ee5, m\u1ed9t m\u00f4 h\u00ecnh ph\u00e1t hi\u1ec7n giao d\u1ecbch gian l\u1eadn c\u00f3 th\u1ec3 \u0111\u01b0\u1ee3c \u0111\u00e0o t\u1ea1o b\u1eb1ng c\u00e1ch s\u1eed d\u1ee5ng h\u1ed3 s\u01a1 c\u00e1c giao d\u1ecbch gian l\u1eadn tr\u01b0\u1edbc \u0111\u00f3.<\/p>\n\n\n\n H\u1ecdc m\u00e1y \u0111\u1ec3 kh\u00e1m ph\u00e1 m\u1eabu (machine learning for pattern discovery<\/strong>)<\/strong> – N\u1ebfu b\u1ea1n kh\u00f4ng c\u00f3 c\u00e1c th\u00f4ng s\u1ed1 \u0111\u1ec3 d\u1ef1a v\u00e0o \u0111\u00f3 \u0111\u01b0a ra d\u1ef1 \u0111o\u00e1n, th\u00ec b\u1ea1n c\u1ea7n t\u00ecm ra c\u00e1c quy lu\u1eadt \u1ea9n trong t\u1eadp d\u1eef li\u1ec7u \u0111\u1ec3 c\u00f3 th\u1ec3 \u0111\u01b0a ra c\u00e1c d\u1ef1 \u0111o\u00e1n c\u00f3 \u00fd ngh\u0129a. M\u00f4 h\u00ecnh n\u00e0y \u0111\u01b0\u1ee3c g\u1ecdi l\u00e0 m\u00f4 h\u00ecnh kh\u00f4ng gi\u00e1m s\u00e1t – b\u1ea1n kh\u00f4ng c\u00f3 s\u1eb5n c\u00e1c ti\u00eau ch\u00ed \u0111\u1ec3 nh\u00f3m c\u00e1c d\u1eef li\u1ec7u v\u1edbi nhau. Thu\u1eadt to\u00e1n ph\u1ed5 bi\u1ebfn nh\u1ea5t \u0111\u01b0\u1ee3c s\u1eed d\u1ee5ng \u0111\u1ec3 ph\u00e2n t\u00edch m\u1eabu trong tr\u01b0\u1eddng h\u1ee3p n\u00e0y l\u00e0 ph\u00e2n c\u1ee5m (clustering).<\/p>\n\n\n\n V\u00ed d\u1ee5, \u0111\u1ec3 thi\u1ebft l\u1eadp m\u1ea1ng l\u01b0\u1edbi, nh\u00e0 m\u1ea1ng \u0111i\u1ec7n tho\u1ea1i di \u0111\u1ed9ng c\u1ea7n \u0111\u1eb7t m\u1ed9t s\u1ed1 tr\u1ea1m thu ph\u00e1t s\u00f3ng \u1edf m\u1ed9t khu v\u1ef1c n\u00e0o \u0111\u00f3. Trong tr\u01b0\u1eddng h\u1ee3p n\u00e0y, nh\u00e0 m\u1ea1ng c\u00f3 th\u1ec3 s\u1eed d\u1ee5ng k\u1ef9 thu\u1eadt ph\u00e2n c\u1ee5m (clustering) \u0111\u1ec3 t\u00ecm c\u00e1c v\u1ecb tr\u00ed \u0111\u1eb7t tr\u1ea1m sao cho t\u1ea5t c\u1ea3 ng\u01b0\u1eddi d\u00f9ng nh\u1eadn \u0111\u01b0\u1ee3c c\u01b0\u1eddng \u0111\u1ed9 t\u00edn hi\u1ec7u t\u1ed1i \u01b0u.<\/p>\n\n\n\n N\u00f3i m\u1ed9t c\u00e1ch kh\u00e1i qu\u00e1t, ph\u00e2n t\u00edch d\u1eef li\u1ec7u (data analysis) bao g\u1ed3m ph\u00e2n t\u00edch m\u00f4 t\u1ea3 (descriptive analysis) v\u00e0 d\u1ef1 \u0111o\u00e1n (predictive analysis) \u1edf m\u1ed9t m\u1ee9c \u0111\u1ed9 nh\u1ea5t \u0111\u1ecbnh. Trong khi \u0111\u00f3, khoa h\u1ecdc d\u1eef li\u1ec7u thi\u00ean nhi\u1ec1u h\u01a1n v\u1ec1 ph\u00e2n t\u00edch d\u1ef1 \u0111o\u00e1n (predictive analysis), ph\u00e2n t\u00edch \u0111\u1ec1 xu\u1ea5t (prescriptive analytics) v\u00e0 h\u1ecdc m\u00e1y (machine learning).<\/p>\n\n\n\n T\u1ea1i sao <\/strong>c\u1ea7n s\u1eed d\u1ee5ng Khoa h\u1ecdc D\u1eef li\u1ec7u?<\/strong><\/p>\n\n\n\n Tr\u01b0\u1edbc kia, d\u1eef li\u1ec7u ch\u00fang ta c\u00f3 ch\u1ee7 y\u1ebfu c\u00f3 c\u1ea5u tr\u00fac v\u00e0 k\u00edch th\u01b0\u1edbc nh\u1ecf, c\u00f3 th\u1ec3 ph\u00e2n t\u00edch \u0111\u01b0\u1ee3c b\u1eb1ng c\u00e1c c\u00f4ng c\u1ee5 BI \u0111\u01a1n gi\u1ea3n. Tuy nhi\u00ean, ng\u00e0y nay h\u1ea7u h\u1ebft d\u1eef li\u1ec7u l\u00e0 kh\u00f4ng c\u00f3 c\u1ea5u tr\u00fac ho\u1eb7c b\u00e1n c\u1ea5u tr\u00fac. \u01af\u1edbc t\u00ednh, v\u00e0o n\u0103m 2020, h\u01a1n 80% d\u1eef li\u1ec7u s\u1ebd l\u00e0 kh\u00f4ng c\u00f3 c\u1ea5u tr\u00fac.<\/p>\n\n\n\n D\u1eef li\u1ec7u n\u00e0y \u0111\u01b0\u1ee3c t\u1ea1o t\u1eeb c\u00e1c ngu\u1ed3n kh\u00e1c nhau nh\u01b0 nh\u1eadt k\u00fd t\u00e0i ch\u00ednh, t\u1ec7p v\u0103n b\u1ea3n, bi\u1ec3u m\u1eabu \u0111a ph\u01b0\u01a1ng ti\u1ec7n, c\u1ea3m bi\u1ebfn v\u00e0 c\u00f4ng c\u1ee5. C\u00e1c c\u00f4ng c\u1ee5 BI \u0111\u01a1n gi\u1ea3n kh\u00f4ng c\u00f3 kh\u1ea3 n\u0103ng x\u1eed l\u00fd kh\u1ed1i l\u01b0\u1ee3ng l\u1edbn v\u00e0 nhi\u1ec1u lo\u1ea1i d\u1eef li\u1ec7u nh\u01b0 v\u1eady. \u0110\u00f3 l\u00e0 l\u00fd do t\u1ea1i sao ch\u00fang ta c\u1ea7n c\u00e1c c\u00f4ng c\u1ee5 v\u00e0 thu\u1eadt to\u00e1n ph\u00e2n t\u00edch ph\u1ee9c t\u1ea1p v\u00e0 n\u00e2ng cao h\u01a1n \u0111\u1ec3 x\u1eed l\u00fd, ph\u00e2n t\u00edch v\u00e0 \u0111\u01b0a ra nh\u1eefng k\u1ebft lu\u1eadn c\u00f3 \u00fd ngh\u0129a.<\/p>\n\n\n\n \u0110\u00f3 kh\u00f4ng ph\u1ea3i l\u00e0 l\u00fd do duy nh\u1ea5t khi\u1ebfn Khoa h\u1ecdc d\u1eef li\u1ec7u tr\u1edf n\u00ean ph\u1ed5 bi\u1ebfn nh\u01b0 v\u1eady. H\u00e3y c\u00f9ng t\u00ecm hi\u1ec3u s\u00e2u h\u01a1n v\u00e0 xem Khoa h\u1ecdc d\u1eef li\u1ec7u \u0111ang \u0111\u01b0\u1ee3c s\u1eed d\u1ee5ng nh\u01b0 th\u1ebf n\u00e0o trong c\u00e1c l\u0129nh v\u1ef1c kh\u00e1c nhau.<\/p>\n\n\n\n B\u1ea1n c\u00f3 mu\u1ed1n hi\u1ec3u c\u00e1c y\u00eau c\u1ea7u ch\u00ednh x\u00e1c c\u1ee7a kh\u00e1ch h\u00e0ng t\u1eeb d\u1eef li\u1ec7u hi\u1ec7n c\u00f3 nh\u01b0 l\u1ecbch s\u1eed duy\u1ec7t web, l\u1ecbch s\u1eed mua h\u00e0ng, tu\u1ed5i v\u00e0 thu nh\u1eadp c\u1ee7a kh\u00e1ch h\u00e0ng kh\u00f4ng? Ch\u1eafc ch\u1eafn l\u00e0 b\u1ea1n \u0111\u00e3 c\u00f3 t\u1ea5t c\u1ea3 d\u1eef li\u1ec7u n\u00e0y tr\u01b0\u1edbc \u0111\u00f3, nh\u01b0ng b\u00e2y gi\u1edd v\u1edbi l\u01b0\u1ee3ng d\u1eef li\u1ec7u phong ph\u00fa v\u00e0 \u0111a d\u1ea1ng, b\u1ea1n c\u00f3 th\u1ec3 \u0111\u00e0o t\u1ea1o m\u00f4 h\u00ecnh hi\u1ec7u qu\u1ea3 h\u01a1n v\u00e0 gi\u1edbi thi\u1ec7u s\u1ea3n ph\u1ea9m cho kh\u00e1ch h\u00e0ng c\u1ee7a m\u00ecnh ch\u00ednh x\u00e1c h\u01a1n. K\u1ebft qu\u1ea3 kinh doanh nh\u1edd \u0111\u00f3 s\u1ebd \u0111\u01b0\u1ee3c c\u1ea3i thi\u1ec7n.<\/p>\n\n\n\n M\u1ed9t v\u00ed d\u1ee5 kh\u00e1c v\u1ec1 vai tr\u00f2 c\u1ee7a Khoa h\u1ecdc d\u1eef li\u1ec7u trong vi\u1ec7c ra quy\u1ebft \u0111\u1ecbnh l\u00e0 xe \u00f4 t\u00f4 t\u1ef1 l\u00e1i. Xe t\u1ef1 l\u00e1i thu th\u1eadp d\u1eef li\u1ec7u tr\u1ef1c ti\u1ebfp t\u1eeb c\u00e1c c\u1ea3m bi\u1ebfn, bao g\u1ed3m radar, camera v\u00e0 tia laser \u0111\u1ec3 t\u1ea1o b\u1ea3n \u0111\u1ed3 v\u1ec1 m\u00f4i tr\u01b0\u1eddng xung quanh. D\u1ef1a tr\u00ean d\u1eef li\u1ec7u n\u00e0y, n\u00f3 \u0111\u01b0a ra c\u00e1c quy\u1ebft \u0111\u1ecbnh nh\u01b0 khi n\u00e0o c\u1ea7n t\u0103ng t\u1ed1c, khi n\u00e0o gi\u1ea3m t\u1ed1c \u0111\u1ed9, khi n\u00e0o c\u1ea7n v\u01b0\u1ee3t, n\u01a1i r\u1ebd – s\u1eed d\u1ee5ng c\u00e1c thu\u1eadt to\u00e1n h\u1ecdc m\u00e1y ti\u00ean ti\u1ebfn.<\/p>\n\n\n\n H\u00e3y xem c\u00e1ch Khoa h\u1ecdc d\u1eef li\u1ec7u c\u00f3 th\u1ec3 \u0111\u01b0\u1ee3c s\u1eed d\u1ee5ng trong ph\u00e2n t\u00edch d\u1ef1 \u0111o\u00e1n. V\u00ed d\u1ee5 trong d\u1ef1 b\u00e1o th\u1eddi ti\u1ebft, d\u1eef li\u1ec7u t\u1eeb t\u00e0u, m\u00e1y bay, radar, v\u1ec7 tinh c\u00f3 th\u1ec3 \u0111\u01b0\u1ee3c thu th\u1eadp v\u00e0 ph\u00e2n t\u00edch \u0111\u1ec3 x\u00e2y d\u1ef1ng m\u00f4 h\u00ecnh. Nh\u1eefng m\u00f4 h\u00ecnh n\u00e0y s\u1ebd kh\u00f4ng ch\u1ec9 d\u1ef1 b\u00e1o th\u1eddi ti\u1ebft m\u00e0 c\u00f2n gi\u00fap d\u1ef1 \u0111o\u00e1n s\u1ef1 xu\u1ea5t hi\u1ec7n c\u1ee7a thi\u00ean tai. N\u00f3 s\u1ebd gi\u00fap b\u1ea1n c\u00f3 nh\u1eefng bi\u1ec7n ph\u00e1p ph\u00f2ng tr\u00e1nh th\u00edch h\u1ee3p v\u00e0 gi\u1ea3m thi\u1ec3u thi\u1ec7t h\u1ea1i\/ th\u01b0\u01a1ng vong do thi\u00ean tai g\u00e2y ra.<\/p>\n\n\n\n Nh\u00e0 khoa h\u1ecdc d\u1eef li\u1ec7u<\/strong> (data scientist) l\u00e0 ai?<\/strong><\/p>\n\n\n\n C\u00f3 nhi\u1ec1u \u0111\u1ecbnh ngh\u0129a v\u1ec1 nh\u00e0 khoa h\u1ecdc d\u1eef li\u1ec7u. Nh\u01b0ng n\u00f3i m\u1ed9t c\u00e1ch \u0111\u01a1n gi\u1ea3n, Nh\u00e0 khoa h\u1ecdc d\u1eef li\u1ec7u l\u00e0 ng\u01b0\u1eddi th\u1ef1c h\u00e0nh ngh\u1ec7 thu\u1eadt c\u1ee7a Khoa h\u1ecdc d\u1eef li\u1ec7u. \u201cNh\u00e0 khoa h\u1ecdc d\u1eef li\u1ec7u\u201d thu th\u1eadp r\u1ea5t nhi\u1ec1u th\u00f4ng tin t\u1eeb c\u00e1c l\u0129nh v\u1ef1c khoa h\u1ecdc v\u00e0 \u1ee9ng d\u1ee5ng kh\u00e1c nhau, kh\u00f4ng ch\u1ec9 l\u00e0 to\u00e1n h\u1ecdc hay th\u1ed1ng k\u00ea.<\/p>\n\n\n\n Nh\u00e0 khoa h\u1ecdc d\u1eef li\u1ec7u l\u00e0m g\u00ec?<\/strong><\/p>\n\n\n\n C\u00e1c nh\u00e0 khoa h\u1ecdc d\u1eef li\u1ec7u l\u00e0 nh\u1eefng ng\u01b0\u1eddi gi\u1ea3i quy\u1ebft c\u00e1c v\u1ea5n \u0111\u1ec1 d\u1eef li\u1ec7u ph\u1ee9c t\u1ea1p b\u1eb1ng chuy\u00ean m\u00f4n c\u1ee7a h\u1ecd trong m\u1ed9t s\u1ed1 l\u0129nh v\u1ef1c khoa h\u1ecdc nh\u1ea5t \u0111\u1ecbnh. H\u1ecd l\u00e0m vi\u1ec7c v\u1edbi m\u1ed9t s\u1ed1 y\u1ebfu t\u1ed1 li\u00ean quan \u0111\u1ebfn to\u00e1n h\u1ecdc, th\u1ed1ng k\u00ea, khoa h\u1ecdc m\u00e1y t\u00ednh, v.v. (m\u1eb7c d\u00f9 h\u1ecd c\u00f3 th\u1ec3 kh\u00f4ng ph\u1ea3i l\u00e0 chuy\u00ean gia trong t\u1ea5t c\u1ea3 c\u00e1c l\u0129nh v\u1ef1c n\u00e0y). H\u1ecd s\u1eed d\u1ee5ng r\u1ea5t nhi\u1ec1u c\u00f4ng ngh\u1ec7 m\u1edbi nh\u1ea5t \u0111\u1ec3 t\u00ecm ra gi\u1ea3i ph\u00e1p v\u00e0 \u0111\u01b0a ra k\u1ebft lu\u1eadn quan tr\u1ecdng \u0111\u1ed1i v\u1edbi s\u1ef1 t\u0103ng tr\u01b0\u1edfng v\u00e0 ph\u00e1t tri\u1ec3n c\u1ee7a t\u1ed5 ch\u1ee9c. Nh\u00e0 khoa h\u1ecdc d\u1eef li\u1ec7u tr\u00ecnh b\u00e0y d\u1eef li\u1ec7u \u1edf d\u1ea1ng h\u1eefu \u00edch h\u01a1n nhi\u1ec1u so v\u1edbi d\u1eef li\u1ec7u th\u00f4 c\u00f3 c\u1ea5u tr\u00fac ho\u1eb7c kh\u00f4ng c\u00f3 c\u1ea5u tr\u00fac ban \u0111\u1ea7u.<\/p>\n\n\n\n B\u00e2y gi\u1edd ch\u00fang ta h\u00e3y th\u1ea3o lu\u1eadn v\u1ec1 BI. C\u00f3 th\u1ec3 b\u1ea1n \u0111\u00e3 nghe n\u00f3i v\u1ec1 Business Intelligence (BI). Khoa h\u1ecdc d\u1eef li\u1ec7u th\u01b0\u1eddng b\u1ecb nh\u1ea7m l\u1eabn v\u1edbi BI. Sau \u0111\u00e2y s\u1ebd l\u00e0 m\u1ed9t s\u1ed1 \u0111i\u1ec3m kh\u00e1c bi\u1ec7t c\u01a1 b\u1ea3n gi\u1eefa 2 l\u0129nh v\u1ef1c n\u00e0y.<\/p>\n\n\n\n Data Science <\/strong>v\u00e0 Business Intelligence (BI)<\/strong> kh\u00e1c nhau nh\u01b0 th\u1ebf n\u00e0o?<\/strong><\/p>\n\n\n\n Business Intelligence (BI) v\u1ec1 c\u01a1 b\u1ea3n ph\u00e2n t\u00edch d\u1eef li\u1ec7u l\u1ecbch s\u1eed \u0111\u1ec3 t\u00ecm ra nh\u1eefng \u0111i\u1ec3m quan tr\u1ecdng nh\u1ea5t nh\u1eb1m m\u00f4 t\u1ea3 xu h\u01b0\u1edbng kinh doanh. V\u1edbi BI, b\u1ea1n s\u1ebd l\u1ea5y d\u1eef li\u1ec7u t\u1eeb c\u00e1c ngu\u1ed3n b\u00ean ngo\u00e0i v\u00e0 n\u1ed9i b\u1ed9, chu\u1ea9n b\u1ecb d\u1eef li\u1ec7u, ch\u1ea1y c\u00e1c truy v\u1ea5n tr\u00ean \u0111\u00f3 v\u00e0 t\u1ea1o trang t\u1ed5ng quan \u0111\u1ec3 tr\u1ea3 l\u1eddi c\u00e1c c\u00e2u h\u1ecfi nh\u01b0 ph\u00e2n t\u00edch doanh thu h\u00e0ng qu\u00fd ho\u1eb7c c\u00e1c v\u1ea5n \u0111\u1ec1 kinh doanh. BI c\u00f3 th\u1ec3 \u0111\u00e1nh gi\u00e1 t\u00e1c \u0111\u1ed9ng c\u1ee7a c\u00e1c s\u1ef1 ki\u1ec7n nh\u1ea5t \u0111\u1ecbnh trong t\u01b0\u01a1ng lai g\u1ea7n.<\/p>\n\n\n\n Khoa h\u1ecdc d\u1eef li\u1ec7u l\u00e0 m\u1ed9t c\u00e1ch ti\u1ebfp c\u1eadn h\u01b0\u1edbng t\u1edbi t\u01b0\u01a1ng lai h\u01a1n, m\u1ed9t c\u00e1ch kh\u00e1m ph\u00e1 v\u1edbi tr\u1ecdng t\u00e2m l\u00e0 ph\u00e2n t\u00edch d\u1eef li\u1ec7u trong qu\u00e1 kh\u1ee9 ho\u1eb7c hi\u1ec7n t\u1ea1i v\u00e0 d\u1ef1 \u0111o\u00e1n k\u1ebft qu\u1ea3 trong t\u01b0\u01a1ng lai nh\u1eb1m \u0111\u01b0a ra quy\u1ebft \u0111\u1ecbnh. N\u00f3 tr\u1ea3 l\u1eddi c\u00e1c c\u00e2u h\u1ecfi m\u1edf v\u1ec1 \u201c\u0111i\u1ec1u g\u00ec\u201d v\u00e0 \u201cl\u00e0m th\u1ebf n\u00e0o\u201d \u0111i\u1ec1u \u0111\u00f3 s\u1ebd x\u1ea3y ra.<\/p>\n\n\n\n C\u00e1c c\u00f4ng vi\u1ec7c c\u1ee5 th\u1ec3 trong Khoa h\u1ecdc d\u1eef li\u1ec7u<\/strong><\/p>\n\n\n\n C\u00e1c c\u00f4ng vi\u1ec7c trong Khoa h\u1ecdc d\u1eef li\u1ec7u c\u00f3 th\u1ec3 chia l\u00e0m 6 giai \u0111o\u1ea1n ch\u00ednh:<\/p>\n\n\n\n Giai \u0111o\u1ea1n 1 \u2014 Kh\u00e1m ph\u00e1: Tr\u01b0\u1edbc khi b\u1ea1n b\u1eaft \u0111\u1ea7u d\u1ef1 \u00e1n, b\u1ea1n c\u1ea7n ph\u1ea3i hi\u1ec3u c\u00e1c th\u00f4ng s\u1ed1 k\u1ef9 thu\u1eadt, y\u00eau c\u1ea7u, m\u1ee9c \u0111\u1ed9 \u01b0u ti\u00ean kh\u00e1c nhau v\u00e0 ng\u00e2n s\u00e1ch c\u1ea7n thi\u1ebft. B\u1ea1n ph\u1ea3i c\u00f3 kh\u1ea3 n\u0103ng \u0111\u1eb7t \u0111\u00fang c\u00e2u h\u1ecfi. \u1ede giai \u0111o\u1ea1n n\u00e0y, b\u1ea1n c\u1ea7n \u0111\u00e1nh gi\u00e1 xem b\u1ea1n c\u00f3 \u0111\u1ee7 c\u00e1c ngu\u1ed3n l\u1ef1c c\u1ea7n thi\u1ebft v\u1ec1 con ng\u01b0\u1eddi, c\u00f4ng ngh\u1ec7, th\u1eddi gian v\u00e0 d\u1eef li\u1ec7u \u0111\u1ec3 h\u1ed7 tr\u1ee3 d\u1ef1 \u00e1n hay kh\u00f4ng. B\u1ea1n c\u0169ng c\u1ea7n \u0111\u1ecbnh h\u00ecnh v\u1ea5n \u0111\u1ec1 kinh doanh v\u00e0 h\u00ecnh th\u00e0nh c\u00e1c gi\u1ea3 thuy\u1ebft ban \u0111\u1ea7u (Initial hypotheses – IH) \u0111\u1ec3 ki\u1ec3m tra.<\/p>\n\n\n\n Giai \u0111o\u1ea1n 2 \u2014 Chu\u1ea9n b\u1ecb d\u1eef li\u1ec7u: Trong giai \u0111o\u1ea1n n\u00e0y, b\u1ea1n c\u1ea7n kh\u00e1m ph\u00e1, l\u00e0m s\u1ea1ch v\u00e0 \u0111\u1eb7t c\u00e1c \u0111i\u1ec1u ki\u1ec7n cho d\u1eef li\u1ec7u tr\u01b0\u1edbc khi x\u00e2y d\u1ef1ng m\u00f4 h\u00ecnh. Ti\u1ebfp \u0111\u1ebfn, b\u1ea1n s\u1ebd th\u1ef1c hi\u1ec7n ETLT (extract, transform, load and transform- tr\u00edch xu\u1ea5t, chuy\u1ec3n \u0111\u1ed5i, t\u1ea3i v\u00e0 chuy\u1ec3n \u0111\u1ed5i) \u0111\u1ec3 c\u00f3 \u0111\u01b0\u1ee3c d\u1eef li\u1ec7u \u0111\u1ea7u v\u00e0o cho m\u00f4 h\u00ecnh.<\/p>\n\n\n\n B\u1ea1n c\u00f3 th\u1ec3 s\u1eed d\u1ee5ng R \u0111\u1ec3 l\u00e0m s\u1ea1ch, chuy\u1ec3n \u0111\u1ed5i v\u00e0 tr\u1ef1c quan d\u1eef li\u1ec7u. T\u1eeb \u0111\u00f3 b\u1ea1n ph\u00e1t hi\u1ec7n ra nh\u1eefng \u0111i\u1ec3m b\u1ea5t th\u01b0\u1eddng v\u00e0 thi\u1ebft l\u1eadp m\u1ed1i quan h\u1ec7 gi\u1eefa c\u00e1c bi\u1ebfn. Sau khi b\u1ea1n \u0111\u00e3 l\u00e0m s\u1ea1ch v\u00e0 chu\u1ea9n b\u1ecb d\u1eef li\u1ec7u, b\u1ea1n c\u00f3 th\u1ec3 b\u1eaft \u0111\u1ea7u th\u1ef1c hi\u1ec7n ph\u00e2n tr\u00ean d\u1eef li\u1ec7u \u0111\u00f3.<\/p>\n\n\n\n Giai \u0111o\u1ea1n 3 \u2014 L\u1eadp k\u1ebf ho\u1ea1ch cho m\u00f4 h\u00ecnh: \u1edf b\u01b0\u1edbc n\u00e0y, b\u1ea1n s\u1ebd x\u00e1c \u0111\u1ecbnh c\u00e1c ph\u01b0\u01a1ng ph\u00e1p v\u00e0 k\u1ef9 thu\u1eadt \u0111\u1ec3 ph\u00e1c th\u1ea3o m\u1ed1i quan h\u1ec7 gi\u1eefa c\u00e1c bi\u1ebfn. Nh\u1eefng m\u1ed1i quan h\u1ec7 n\u00e0y s\u1ebd \u0111\u1eb7t c\u01a1 s\u1edf cho c\u00e1c thu\u1eadt to\u00e1n m\u00e0 b\u1ea1n s\u1ebd tri\u1ec3n khai trong giai \u0111o\u1ea1n ti\u1ebfp theo. B\u1ea1n s\u1ebd \u00e1p d\u1ee5ng Ph\u00e2n t\u00edch d\u1eef li\u1ec7u kh\u00e1m ph\u00e1 (Exploratory Data Analytics – EDA) b\u1eb1ng c\u00e1ch s\u1eed d\u1ee5ng c\u00e1c c\u00f4ng th\u1ee9c th\u1ed1ng k\u00ea v\u00e0 c\u00f4ng c\u1ee5 tr\u1ef1c quan h\u00f3a kh\u00e1c nhau.<\/p>\n\n\n\n M\u1ed9t s\u1ed1 c\u00f4ng c\u1ee5 \u0111\u1ec3 l\u1eadp k\u1ebf ho\u1ea1ch m\u00f4 h\u00ecnh:<\/p>\n\n\n\n Tuy c\u00f3 nhi\u1ec1u c\u00f4ng c\u1ee5 kh\u00e1c nhau nh\u01b0ng R l\u00e0 c\u00f4ng c\u1ee5 \u0111\u01b0\u1ee3c s\u1eed d\u1ee5ng ph\u1ed5 bi\u1ebfn nh\u1ea5t.<\/p>\n\n\n\n Giai \u0111o\u1ea1n 4 \u2014 X\u00e2y d\u1ef1ng m\u00f4 h\u00ecnh: Trong giai \u0111o\u1ea1n n\u00e0y, b\u1ea1n s\u1ebd x\u00e2y d\u1ef1ng c\u00e1c b\u1ed9 d\u1eef li\u1ec7u cho m\u1ee5c \u0111\u00edch \u0111\u00e0o t\u1ea1o v\u00e0 th\u1eed nghi\u1ec7m. T\u1ea1i \u0111\u00e2y, b\u1ea1n c\u1ea7n xem x\u00e9t li\u1ec7u c\u00e1c c\u00f4ng c\u1ee5 hi\u1ec7n c\u00f3 c\u1ee7a b\u1ea1n c\u00f3 \u0111\u1ee7 \u0111\u1ec3 ch\u1ea1y c\u00e1c m\u00f4 h\u00ecnh hay kh\u00f4ng, hay c\u1ea7n m\u1ed9t m\u00f4i tr\u01b0\u1eddng m\u1ea1nh m\u1ebd h\u01a1n (nh\u01b0 x\u1eed l\u00fd nhanh v\u00e0 song song). B\u1ea1n s\u1ebd ph\u00e2n t\u00edch c\u00e1c k\u1ef9 thu\u1eadt h\u1ecdc m\u00e1y kh\u00e1c nhau nh\u01b0 ph\u00e2n lo\u1ea1i, li\u00ean k\u1ebft v\u00e0 ph\u00e2n c\u1ee5m \u0111\u1ec3 x\u00e2y d\u1ef1ng m\u00f4 h\u00ecnh.<\/p>\n\n\n\n C\u00e1c c\u00f4ng c\u1ee5 ph\u1ed5 bi\u1ebfn \u0111\u1ec3 x\u00e2y d\u1ef1ng m\u00f4 h\u00ecnh:<\/p>\n\n\n\n Giai \u0111o\u1ea1n 5 \u2014 V\u1eadn h\u00e0nh: Trong giai \u0111o\u1ea1n n\u00e0y, b\u1ea1n tr\u00ecnh b\u00e0y c\u00e1c b\u00e1o c\u00e1o, b\u1ea3n t\u00f3m t\u1eaft, code v\u00e0 t\u00e0i li\u1ec7u k\u1ef9 thu\u1eadt. Ngo\u00e0i ra, \u0111\u00f4i khi b\u1ea1n c\u00f2n c\u00f3 th\u1ec3 th\u1ef1c hi\u1ec7n m\u1ed9t d\u1ef1 \u00e1n th\u1eed nghi\u1ec7m trong m\u00f4i tr\u01b0\u1eddng s\u1ea3n xu\u1ea5t th\u1ef1c. \u0110i\u1ec1u n\u00e0y s\u1ebd cung c\u1ea5p cho b\u1ea1n m\u1ed9t b\u1ee9c tranh r\u00f5 r\u00e0ng v\u1ec1 hi\u1ec7u su\u1ea5t v\u00e0 c\u00e1c r\u00e0ng bu\u1ed9c li\u00ean quan kh\u00e1c \u1edf quy m\u00f4 nh\u1ecf tr\u01b0\u1edbc khi tri\u1ec3n khai \u0111\u1ea7y \u0111\u1ee7.<\/p>\n\n\n\n Giai \u0111o\u1ea1n 6 \u2014 Truy\u1ec1n \u0111\u1ea1t k\u1ebft qu\u1ea3: t\u1ea1i \u0111\u00e2y, b\u1ea1n ph\u1ea3i \u0111\u00e1nh gi\u00e1 xem b\u1ea1n c\u00f3 th\u1ec3 \u0111\u1ea1t \u0111\u01b0\u1ee3c m\u1ee5c ti\u00eau m\u00e0 b\u1ea1n \u0111\u00e3 l\u00ean k\u1ebf ho\u1ea1ch trong giai \u0111o\u1ea1n \u0111\u1ea7u hay kh\u00f4ng. V\u00ec v\u1eady, trong giai \u0111o\u1ea1n cu\u1ed1i, b\u1ea1n x\u00e1c \u0111\u1ecbnh t\u1ea5t c\u1ea3 c\u00e1c ph\u00e1t hi\u1ec7n ch\u00ednh, th\u00f4ng b\u00e1o cho c\u00e1c b\u00ean li\u00ean quan v\u00e0 x\u00e1c \u0111\u1ecbnh xem k\u1ebft qu\u1ea3 c\u1ee7a d\u1ef1 \u00e1n l\u00e0 th\u00e0nh c\u00f4ng hay th\u1ea5t b\u1ea1i d\u1ef1a tr\u00ean c\u00e1c ti\u00eau ch\u00ed \u0111\u00e3 ph\u00e1t tri\u1ec3n trong Giai \u0111o\u1ea1n 1. Khi th\u1ebf gi\u1edbi b\u01b0\u1edbc v\u00e0o th\u1eddi \u0111\u1ea1i d\u1eef li\u1ec7u l\u1edbn (big data), nhu c\u1ea7u l\u01b0u tr\u1eef c\u0169ng t\u0103ng l\u00ean. \u0110\u00f3 […]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"nf_dc_page":"","om_disable_all_campaigns":false,"_monsterinsights_skip_tracking":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[3],"tags":[],"class_list":["post-71","post","type-post","status-publish","format-standard","hentry","category-kien-thuc"],"jetpack_sharing_enabled":true,"jetpack_featured_media_url":"","_links":{"self":[{"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=\/wp\/v2\/posts\/71","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=71"}],"version-history":[{"count":3,"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=\/wp\/v2\/posts\/71\/revisions"}],"predecessor-version":[{"id":84,"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=\/wp\/v2\/posts\/71\/revisions\/84"}],"wp:attachment":[{"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=71"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=71"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/datasciencebasic.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=71"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"\"](\"https:\/\/i0.wp.com\/datasciencebasic.com\/wp-content\/uploads\/2021\/03\/what-is-ds.png?resize=567%2C169&ssl=1\")
<\/figure>\n\n\n\n
<\/p>\n","protected":false},"excerpt":{"rendered":"