Tag Archives: Predictive Analytics

Qlik publishes new release “Qlik Sense June 2017”

Qlik Sense June 2017BI provider Qlik has published a new release of Qlik Sense. The latest version contains various interesting features such as better charts and visualization options, advanced analytics functions, an iOS client, visual data profiling and more. The version is named “Qlik Sense June 2017” in line with the new release cycle announced by Qlik.

Already at this year’s Qonnections, Qlik announced that in future new software releases of equal weight and scope will be published every ten weeks – signifying an end to major and minor releases. To clarify this, they replaced the numerical name with a description containing the month and year of the respective release. Qlik Sense 4.0 thus became Qlik Sense June 2017. The new cloud-like release cycle and changed versioning should significantly simplify upgrading to higher versions.

New features for data processing and visualization

Qlik Sense June 2017 contains various new features, particularly in data processing and visualization options. Users, among other things, benefit from improved data preparation and provision without complex scripting. With visual data profiling Qlik Sense June 2017, e.g., supports better understanding data from the get go by automatically generating visualizations for the distribution of data already when loading. Furthermore Data binning allows for the grouping of numeric data (by size etc.) to better analyze relevant information. Using a new type of table concatenation, table views with different fields and field names can additionally be connected. This makes it possible to concatenate differently created data sets (e.g., from different organizations). The data preparation functions are introduced in the next What’s New video:

Qlik Sense June 2017 - Data Preparation

Continue reading Qlik publishes new release “Qlik Sense June 2017”

Optimise Business Processes with Machine Learning and Dynamics NAV 2017

Machine Learning with Dynamics NAV 2017The unstoppable progress of digital transformation not only changes the way of living and working of private individuals, but also presents big challenges to companies on a regular base. An efficient way of dealing with the increasingly rapid transformation are modern technologies and approaches such as machine learning.

In order to stay at the pulse of the times, executives should prepare themselves in time for the continuing digitalisation of the business world. This digitalisation also offers companies numerous advantages that not only facilitate administrative processes and general everyday operations, but also secure competitiveness in the long term. For example:

  • Creation of flexible work forms
  • Service optimisation through online support
  • Cross-location working
  • Streamlining administration
  • Outsourcing administrative processes

In order to strengthen or even improve ones’ own position in the competition, to increase productivity, quality and sales, and to save money by a more efficient use of resources, the use of machine learning is recommended. Machine learning is not a finished software product, but an individualisable learning and optimisation process. Using machine learning, algorithms are developed with the aim of analysing historical data automatically and identifying patterns therein. The results can finally be used as predictions and thus as a basis for future business decisions or to solve problems.

Efficient machine learning with Microsoft Dynamics NAV

Using Microsoft Azure Machine Learning in combination with the data source Microsoft Dynamics NAV 2017 is worthwhile for the introduction of machine learning into ones’ own company. The advantages for the companies include the following:

  • Assess the sales potential
    Efficient business management with the help of “learning” sales forecasts based on historical data.
  • Prompt overview of stock shortages
    Timely overview of inventory developments based on current stock bookings and dynamic sales forecasts.
  • Decision optimisation
    Realistic predictions as solid decision-making bases by means of various machine-learning algorithms and continuous result comparisons.
  • Transparency of predictive quality
    Evaluation of the calculated prediction results in everyday practice. Successive optimisation of prediction quality through exclusion or weighting of the predictions obtained.

Machine learning is used wherever large quantities of data are analysed and compactede into business or scientifically quantity and value indicators. According to experts, machine learning will be ready for the market and business-relevant across the board in two to five years. It is worth getting started with it now.

Get an overview over Microsoft Azure Machine Learning in the following video:

Microsoft Azure Machine Learning

Microsoft Dynamics NAV 2017 – Cortana Intelligence

Thanks to the option of expanding the NAV environment by means of various extensions, Dynamics NAV 2017 users can now also view sales and inventory development forecasts using machine learning.

Dynamics NAV can be expanded with an extension that gives the possibility to acquire knowledge of potential future sales from past data and a clear overview of expected stocks and sales development. Microsoft runs this extension under the name “Cortana Intelligence”. The Dynamics NAV extensions have no functional or content-related connection to the “Cortana Intelligence” concept in Power BI.

The system uses historical data for forecasts and helps the user to manage company stocks more efficient. This supports the supply management and ensures the customers satisfaction by guaranteeing that frequently purchased goods are consistently in stock.

Inventory Forecast in Microsoft Dynamics NAV 2017
Inventory Forecast in Microsoft Dynamics NAV 2017

In order to achieve optimal forecasts all the time, the extension uses a trained algorithm within the cloud service Azure Machine Learning, determines a corresponding result using the training data and finally issues this data to the user.

The performance range of the extension “Sales and Inventory Forecast” in Dynamics NAV 2017 covers the following functional areas on the base of historical data:

  • Forecast within the cash flow module
  • Inventory forecast and direct creation of purchase orders
  • Sales forecast

In addition to the existing functionality, Dynamics NAV 2017 also offers developers access to the trained algorithm within Azure Machine Learning. In this way, forecasts for the future can also be accessed on the base of further historical data.