International Business Machines Corporation

Alex Melament

Edward Vitkin

Yardena Lia Peres

Igor Kostirev-Kronos

Gal A Shachor

Alexey Roytman

Uri Shani

Bazarbek Uatay

Igor Kostirev

Time-series data uncertainty reduction can include generating an initial accumulated signal based on an event classifier prediction score. The event classifier prediction score can be generated by an event predicator based on time-series data and can correspond to a probability that a target event occurs. Signal leakage can be imposed on the initial accumulated signal. Additionally, an alert can be generated in response to determining that the initial accumulated signal is greater than an alert threshold.

Time-series data uncertainty reduction

A method comprising using at least one hardware processor for: receiving multiple data clusters, each comprising one or more path variations of a process performed with respect to multiple subjects, wherein each of the path variations comprises multiple stages of the process, and wherein at least some of the stages, each comprises one or more parameters; constructing a visualization template representative of the path variations, wherein the visualization template comprises multiple nodes, each node having one or more graphical attributes, wherein each node representative of a corresponding stage; assigning each of the graphical attributes of each of the nodes to a corresponding parameter of the corresponding stage; and visualizing one or more differences between the data clusters by generating at least one instance of the visualization template, the instance being representative of and corresponding to at least two of the data clusters, wherein each of the at least one instance is representative of and corresponding to at least one of the data clusters, and wherein in the at least one instance, each of the assigned one or more graphical attributes of each node represent a value of the corresponding parameter, the value relating to the corresponding stage of the at least one corresponding data cluster.

Visual comparison of data clusters

An inexpensive storage system is disclosed along with methods of managing such a system. In one preferred embodiment, the system includes a high performance high reliability storage medium configured for initial storage of data, a low performance high reliability storage medium configured for backup of data initially stored on the high performance high reliability storage medium, and a high performance low reliability storage medium, configured to receive data transferred from the high performance high reliability storage medium, after the data has been backed up on the low performance high reliability storage medium. This significantly reduces the cost of the system without substantially comprising performance. Reliability is likewise maintained owing to the high reliability backup.

Inexpensive reliable computer storage via hetero-geneous architecture and a staged storage policy

Growing community of inventors

Kiriat Mozkin, Israel

Alex Melament