Determining Recommendations In Data Analysis

• US 20150046439A1
• Filed: 09/03/2014
• Published: 02/12/2015
• Est. Priority Date: 08/06/2013
• Status: Abandoned Application

• Current Assignee: GlobalFoundries, Inc.
• Original Assignee: International Business Machines Corporation
• Inventors: Gokhale, Parag S., Grosset, Robin N., Malviya, Rajanikant, Mittal, Amit
• Application Number:
  US14/475,602
• Publication Number:
  US 20150046439A1
• Time in Patent Office: Days
• US Class Current: 707/723
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06Q 10/063 Operations research, analysis or management

Abstract

Embodiments of the present invention disclose a method, computer program product, and system for determining recommendations in data analysis. A computer identifies an analysis step currently being performed in a data analysis. The computer identifies data points corresponding to the identified analysis step currently being performed and one or more previous analyses. The computer determines a distance between the data points corresponding to the identified analysis step currently being performed and each of the one or more previous data analyses utilizing a distance computing algorithm. The computer determines a ranking of the one or more previous data analyses corresponding to the determined distances between the data points corresponding to the identified analysis step currently being performed and each of the one or more previous data analyses.

Claims

1. A method for determining recommendations in data analysis, the method comprising:

a computer identifying an analysis step currently being performed in a data analysis;

the computer identifying data points corresponding to the identified analysis step currently being performed and one or more previous analyses;

the computer determining a distance between the data points corresponding to the identified analysis step currently being performed and each of the one or more previous data analyses utilizing a distance computing algorithm; and

the computer determining a ranking of the one or more previous data analyses corresponding to the determined distances between the data points corresponding to the identified analysis step currently being performed and each of the one or more previous data analyses.

2. The method of claim 1, further comprising:

the computer determining recommendations in the data analysis utilizing the determined ranking of the one or more previous data analyses, wherein the recommendations include possible next analytical steps that correspond to the one or more previous data analyses.

3. The method of claim 1, wherein the one or more previous data analyses are stored previous steps in data analysis that include parameters corresponding to performing data analysis steps of the one or more previous data analyses.

4. The method of claim 1, wherein the computer identifying data points corresponding to the identified analysis step currently being performed and one or more previous analyses, comprises:

the computer indexing data included in the one or more previous analyses utilizing text indexing software, wherein the indexed data includes attributes of the one or more previous analyses and data corresponding to the attributes;

the computer determining a numerical representation of the indexed data, wherein data of the one or more previous analyses is represented by a numerical data point; and

the computer determining a numerical representation of data points of the identified analysis step currently being performed and one or more previous analyses.

5. The method of claim 1, further comprising:

the computer receiving an indication of a distance computing algorithm to utilize in determining distance between the data points corresponding to the identified analysis step currently being performed and each of the one or more previous data analyses,

wherein the received distance computing algorithm is an algorithm that is utilized to determine distance between data points in a multidimensional space.

6. The method of claim 1, further comprising

the computer receiving an indication of preferences for determining the ranking of the one or more previous data analyses,

wherein the received preferences include an indication of factors to be utilized in determining the ranking of the one or more previous data analyses.