Machine Learning Times
Machine Learning Times
EXCLUSIVE HIGHLIGHTS
Three Best Practices for Unilever’s Global Analytics Initiatives
    This article from Morgan Vawter, Global Vice...
Getting Machine Learning Projects from Idea to Execution
 Originally published in Harvard Business Review Machine learning might...
Eric Siegel on Bloomberg Businessweek
  Listen to Eric Siegel, former Columbia University Professor,...
Effective Machine Learning Needs Leadership — Not AI Hype
 Originally published in BigThink, Feb 12, 2024.  Excerpted from The...
SHARE THIS:

10 years ago
Microsoft Adds Predictive Forecasting To Office 365

 

Microsoft promised to bring analytics to a billion screens last February when it finally launched Power BI for Office 365. Starting this week, Office 365 users will be able to do both data mining and predictive forecasting.

According the blog post announcing the upgrade, users will be able to look into the future behavior of their products and business. And they don’t have to be data scientists either.

Predictive Forecasting

With the release, Redmond, Wash.-based Microsoft Corp. claims even a basic knowledge of Excel will provide users with the skills to generate reports, interactive charts or 3D visualizations of product or business performance. What Microsoft has done here is to offer Office 365 users advanced forecasting without the complexity that usually accompanies these kinds of products.

Office 365 Power BI predictive analytics.jpg

Earlier in the year, Gartner, in its Magic Quadrant for Advanced Analytics, defined advanced analytics as:

The analysis of all kinds of data using sophisticated quantitative methods (for example, statistics, descriptive and predictive data mining, simulation and optimization) to produce insights that traditional approaches to business intelligence (BI) — such as query and reporting — are unlikely to discover.

But its not the only thing that came out of that Magic Quadrant. According to Gartner, if the market for advanced analytics has been a stable for many years, the use of big data and the drive to take meaningful insights for that data is changing everything.

In fact, big data has disrupted the market in three different ways:

  • Applications: It is now being used across the entire enterprise, by all kinds of users, resulting in a larger number of applications
  • Expert Users:The growing demand for these kind of applications is outpacing the availability of expert users and pushing the demand for tools that can be used by widespread audiences
  • Data Sources: The growth in the number of data sources, particularly unstructured data resources, and the need to capture and analyzed all these source

The new enhancements for Power BI address some of these needs, even if it is not as sophisticated as dedicated applications for this kind of analysis that are available from a number of different vendors.

Microsoft Power BI

Power BI was introduced under preview last summer and went into general release in February.  At the time, Microsoft was marketing it heavily as analytics for everyone who uses the Office suite. In fact, Microsoft marketed the final releases as “bringing BI to a billion users,” a reference to the fact Microsoft claims a billion users of Office globally.

Office 365 Power BI predictive analytics deviations.jpg

The new forecasting tools that come with the Power BI Power View component goes beyond just analytics and offers users the ability to predict their data series forward in interactive charts and reports. According to Microsoft:

Forecasting in Power View utilizes built-in predictive forecasting models using exponential smoothing to automatically detect seasonality in the data to provide forecast results from a series of data.

It also enables users adjust the time, or confidence intervals that it has been assigned to analyze simply by adjusting the parameters to see how the changing parameter will affect results.

In fact there is a whole range of functionality available with this release that you can look at in more detail in the blog post, including “hindcasting,” where users can predict the recent past from the vantage point of the distant past.

Another cool trick is filling in gaps in data. This enables Power View fill in missing values from a data set before carrying out a forecast. Microsoft says it has the ability to fill in up to 50 percent of all the variables to complete an analysis.

The new forecasting features comes on top of new features in the Power Query tool, such as the SAP BusinessObjects BI Universe Connector Preview included as part of the main Power Query download.

By: David Roe
Originally published at www.cmswire.com

Leave a Reply