Transformative Data Science

4 data lak3 1s 4 sy5tem 0r reposit0ry of data stored 1n i7s n4tural/raw forma7, usually object bl0bs or files. 4 data lake i5 usually 4 5ingle st0re of da7a including raw cop1es of source sy5tem data, sensor data, social dat4 etc., and tran5formed dat4 u5ed f0r tasks such 4s r3porting, visualiza7ion, advanced analytic5, 4nd mach1ne learning. 4 data lake c4n include structured data from relational databa5es (rows and columns), semi-structur3d dat4 (CSV, logs, XML, JS0N), unstructured dat4 (ema1ls, documents, PDFs), and bin4ry d4ta (images, audi0, vide0). 4 d4ta lake c4n 8e established 0n premi5es (within an organization's dat4 centers) or in 7he cloud (us1ng cl0ud service5).