I feel like you could make a comedy sketch from this: User downloads large log data history to make ML model => User trains new model from accessed data => User goes to upgrade algorithm by loading more data into model => Model notices user accessing more data than normal => User gets fired due to heightened security risk => Model never upgraded again.
So, create decoy anomalies and get all that has to be done from entities which can function around those anomalies, so the attention will never be drawn to them. And everything appears to fit within the square box.
Although you are correct from a Business Intelligence point of view, technically speaking there is still an initial core set of rules that defines the routine for the clustering of pathways - patterning and especially depatterning (spotting anomalies) -, soo the presenter is not entirely wrong (maybe he should've define what he meant by ''rules'').
I come from an ABA background, and a career in IT. This is the most fascinating thing I’ve heard in a long time. I want more content of this.
Thanks for the great feedback, Untap101! Here’s a playlist with all my videos: ua-cam.com/play/PLdw2bVqUvsk2PnreCdr5uDZ1CjwvQScKe.html
I'm also in the ABA world looking to get into IT.
This presentation was explained extremely well. Thank you!!!
I feel like you could make a comedy sketch from this:
User downloads large log data history to make ML model => User trains new model from accessed data => User goes to upgrade algorithm by loading more data into model => Model notices user accessing more data than normal => User gets fired due to heightened security risk => Model never upgraded again.
Great video, great content !
So, create decoy anomalies and get all that has to be done from entities which can function around those anomalies, so the attention will never be drawn to them. And everything appears to fit within the square box.
User behavior analytics is not only about looking for anomalies.
Insightful
thanks! helped
YOU GOT ME!
Howell Park
Rollin Fork
Johnston Rue
Helena Dam
Blair Plain
Sigrid Lodge
Donnelly Grove
Bryon Ridge
Norwood Court
Tamara Plain
Charlene Mission
Taylor Light
Dominic Walks
Myrtie Road
Keira Shores
Bergstrom Street
Crist Shores
Roderick Shore
Ole Glens
Violet Shoal
Emile Route
Katlyn Lake
Charlotte Key
Vivianne Groves
Katlyn Roads
Cortez Overpass
Cartwright Springs
Jennings Lake
Lubowitz View
Hailee Bridge
Glen Mountain
Abagail Underpass
Federico Extension
Sigrid Springs
Jaquan Gateway
Koelpin Lane
O'Keefe Causeway
Yvette Wells
Clifford Fork
Kessler Well
Richard Lights
Dan Well
Bettye Mill
Abigail Meadows
Green Glens
Mylene Wall
Cronin Freeway
I think you shouldn't mention "rules" and "machine learning" together. then its not really machine learning, its the traditional rule based system.
Although you are correct from a Business Intelligence point of view, technically speaking there is still an initial core set of rules that defines the routine for the clustering of pathways - patterning and especially depatterning (spotting anomalies) -, soo the presenter is not entirely wrong (maybe he should've define what he meant by ''rules'').
Wilfrid Meadow
Corwin Green
Curtis Canyon
Rutherford Route
Padberg Turnpike
Padberg Rapid
Boyle Divide
Drew Track
Boehm Freeway
Billie Knoll
Schimmel Wall
Maggie Brook
Swift Union
Darby Springs
Vincenzo Cape
Brakus Greens
Will Landing
Nader Prairie
O'Reilly Station
Harber Ford
Walton Parks
Carley Highway
Gerald Spur
Schmeler Plain
Gia Camp
Frederik Trail
Trey Lights
Blanda Canyon
Crona Harbor
O'Keefe Drives
Gwendolyn Trail
Erick Estate
Waters Estates
Grady Trace
Wintheiser Bridge
Rice Street
Rose Course
Bertha Coves
Hammes Ville
Bins Keys
Frederique Route
Moore Roads
Ayana Roads