Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Method and System for Analytic Model and Data Visualizatoin on Mobile Devices

IP.com Disclosure Number: IPCOM000236178D
Publication Date: 2014-Apr-11
Document File: 3 page(s) / 124K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosured herein is method and a configurable system for analytic model and data visualization on mobile devices. A mobile database on mobile device is used to keep analytic model and a sub-set of data. Analytic model and data are tailored based on user's configuration on the display. Full set of data is saved on a server that mobile device has connection to. Scoring can be configured to happen on either server, or mobile device. Analytic model, together with scoring result are visualized based on user defined criteria.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 01 of 3

Method and System for Analytic Model and Data Visualizatoin on Mobile Devices

The architecture of the system is shown in the following figure (fig 1). The system, deployed on a mobile device, is composed of a data visualization module, a model visualization module, a scoring component module and a mobile database. The data visualization module retrieves data by generating SQL query on the fly, based on user-defined criteria. The model visualization module tailor the model to fit the screen, based on user selection. The scoring component module is a set of dynamic linked libraries with embedded analytic capability to build more and score data. Both mobile database and back-end database shares the same table definition that holds analytic models. The data set that user is currently working on is loaded as an in-memory table for optimized query performance.

Fig 1. System Architecture

1


Page 02 of 3

In the preferred embodiment, a user navigates through a list of analytic models from mobile device. In alternative embodiment, analytic models and data visualization could happen on any devices that has limited access to the back -end server, the device shall be equipped with computing facility and storage media for processing scoring/visualization and persistence of model and data.

In the preferred embodiment, as the user open one of the models on the screen, the model is loaded from back-end server to mobile database on mobile device. The user picks up several factors that is more interesting, and picks up several fields from definition of a data set that is going to be scored using the model. As the model is displayed on the screen, only part of the model that meet selected factors are shown. In another table, a sample of data containing selected fields is shown. The user changes some of the parameters of the model, and chooses to score only part of the data that is synchronized and saved in mobile database. The data containing scoring result (as shown in another table) is refreshed with the new scoring result.

In the preferred embodiment, scoring component module contains third party analytic libraries, which can be invoked via User-Defined Function (UDF) call by specifying model ID, predictors and model parameters. In alternative embodiment, scoring component module may be delivered as DBMS built-in analytic functions, callable in either UDFs, or Stored Procedures.

The following details how the system works:
User navigates through a list of analytic models on mobile device


1.

The list of models is retrieved from back-end server


a.

The list of models is saved in a table in mobile database installed on mobile device

b.

User opens one of the models on the list


2.

System load the model property to the screen


3.

System checks whether the model exis...