Image source: https://powerplatform.microsoft.com/en-us/
What is Microsoft Power Platform
Microsoft Power Platform is an integrated application platform, consisting of Power BI, PowerApps and Microsoft Flow.. This combination of Microsoft elements allows developers to create powerful enterprise solutions.
It was presented in 2018 as Microsoft's big bet, to complete its offer of cloud storage services in Azure, productivity and office automation with Microsoft 365 and business management through Dynamics 365.
What is Microsoft Power BI
Microsoft Power BI is the leading data analysis, business intelligence & Big Data tool on the market, according to Forrester. It allows you to connect to various data sources, analyse them in real time, visualise the results in a very attractive way and share them with the entire organisation, or publish them on the Internet.
In the following article you will find all the information about the advantages of Power BI, types of licenses and which one is best suited to each type of company: Read more
What is PowerApps
PowerApps is a Microsoft tool for easily and quickly creating enterprise mobile applications.
What is Microsoft Flow
Microsoft Flow is a tool for creating agile workflows to automate tasks and business processes.
Image source: https://powerplatform.microsoft.com/en-us/
These three applications integrate seamlessly with the Microsoft Power Platform.
For example, we can embed interactive business intelligence reports from Power BI within Power Apps. In this way they share both data input and output. And another example in reverse, is the possibility of using Power Apps to get user information and embed it within Power BI reports or dashboards. Microsoft Flow would come into the picture when we want to create a workflow. For example: Within the Power Apps designer you can link workflows developed with Microsoft Flow.
In conclusion, with the Microsoft Power Platform, end-to-end solutions can be developed, ranging from user data entry, via a Power Apps application; to processing that data, via Microsoft Flow; and finally, to real-time reporting, via Microsoft Power BI.
Data storage in Power Platform
In relation to the issue of data storage and management, these components of the Microsoft Power Platform) are based on the Common Application Data Service, or Common Data Model (CDM: Common Data Model). In terms of data connectivity, all three applications, Power BI, PowerApps and Microsoft FlowThe data connectors and gateways are shipped with a large number of local data connectors and gateways, as can be seen in the following graphic:
Power Platform user groups
There are several groups of advanced and professional Power Platform users in Spain. They are online communities where you can get updated information on the latest news about the platform and where you can share knowledge and experiences in Power BI, PowerApps or Microsoft Flow. These are the most important groups: PUGSpain, PowerBIUG PowerAppsUG and FlowUG
Any professional or advanced user of Power Platform or its applications can join these communities and enjoy their benefits free of charge.
Power Platform and Artificial Intelligence:
Source: https://powerapps.microsoft.com/en-us/blog/introducing-ai-builder-for-powerplatform/
What is Microsoft AI Builder
Microsoft AI Builder was presented last June at the Microsoft Business Applications Summit (MBAS) in Atlanta, and is available in draft form.
AI Builder is a new capability of the Power Platform, which uses artificial intelligence and machine learningto create smarter, more effective PowerApps and Microsoft Flow automations. It enables predictive analytics and machine learning, without the need to write code.
AI Builder allows applications to be built on PowerApps that make use of artificial intelligence. It also enables the consumption of cognitive services in the automation of Microsoft Flow flows. These cognitive services use artificial intelligence to perform basic vision, speech, language, search and decision algorithms.
Types of AI Builder Artificial Intelligence solutions
Source: https://docs.microsoft.com/en-us/powerapps/use-ai-builder
It includes five fundamental AI models:
1) Object detection and/or image recognition
It allows physical objects to be recognised. An example of an application would be inventory management, where it would allow products to be recognised in order to update the stock.
2) Binary classification
Ability to predict yes/no or true/false outcomes based on historical data, identifying patterns and comparing them to current data. For example, it can be used to predict whether a customer will make a purchase or abandon based on previous patterns of behaviour.
3) Forms processing (OCR)
Ability to extract data from paper or PDF forms and store it in the Data Centre. For example, it can be used to identify, download and store data from invoices that we receive in paper or PDF, avoiding having to manually enter the data. This function can be used in both PowerApps and Microsoft Flow.
4) Text classification
Allows you to classify text and documents, with greater accuracy and efficiency. It can classify, group and categorise large volumes of data. It can be used, for example, to automate the analysis of survey results.
5) Business Card Reader
Allows data to be extracted from a business card and stored on the CDS as contacts.
How to use AI Builder
1. First of all, you have to choosing the artificial intelligence model
The most suitable of the above is the one that best suits your needs.
2. Connecting to CDS, Common Data Service data
.
3. Adapting the model
of AI selected to the business problem we want to solve.
4. Training the model
The resulting artificial intelligence : i.e. test it several times.
5. Use the AI model created to develop applications with Power Platform, or automate tasks with Microsoft Flow..
Microsoft Power BI Dataflows, Machine Learning & Cognitive Services
What is Machine Learning
Machine Learningis a branch of artificial intelligence that allows machines to learn autonomously. It is a method of data analysis that automates the building of analytical models, i.e. it allows computers to automatically learn how to use existing data to predict future trends and outcomes.
Azure Machine Learning Studio is a collaborative tool from Microsoft in the cloud that allows you to create, train and deploy machine learning solutions without programming. Machine Learning Studio publishes machine learning models as web services, which can be easily used in applications or tools such as Power BI. It is very easy to use as it does not require programming to create the predictive analytics model, just plug in data sets and analytics modules, which are dragged onto an interactive canvas where they are connected. It can be run several times to test it and once ready, it becomes a predictive experiment and can be published as a web service.
AutoML Power BI
Automated Machine Learning (AutoML) Enables business analysts to create machine learning models without programming. Train, validate, and use machine learning models directly in Power BI. It includes a simple experience to create a new machine learning model where analysts can use their data streams to specify the input data to train the model. The service automatically extracts the most relevant features, selects an appropriate algorithm, and optimises and validates the machine learning model. After a model is trained, Power BI automatically generates a report including the validation results that explains the model's performance. Finally, the model can be invoked on new or updated data in the data stream.
To create a machine learning model in Power BI, you must first create a data flow, for data with historical outcome information, which is used to train the predictive analytics model. For more information on how to set up the dataflow, see Self-Service Data Preparation in Power BI.
Dataflows
Image source: https://docs.microsoft.com/es-es/power-bi/service-dataflows-overview
Dataflows is a functionality within Power BI Service, for performing ETL processes in the cloud, which allows you to prepare data for further analysis, easily and quickly. Dataflows creates a collection of entities, called a dataflow, which you can then use to connect to data sources. With just a few clicks, the entire process is done. ETL, Extract, Transform and LoadExtract data from multiple sources or origins, cleanse, transform, enrich and load it into another database, or datawarehouse, for further analysis.
Dataflows connects to Common Data Service in Azure Data Lake Storage Gen2, where the data is stored (available only in Azure Data Lake Gen2). Power BI Pro and Premium). To create a Dataflow you have to access directly to the Power BI online service, via https://app.powerbi.com
Once the data flows are ready, with AutoML we can create the machine learning models, but only with the version Power BI Premium & Embedded
Steps to create a predictive analytics model in Power BI
Source : https://docs.microsoft.com/es-es/power-bi/service-machine-learning-automated
1. Select the databases
2. Choosing a model type
Train the model and generate a performance report.
4. Improve the model: Evaluate and retest until it is valid.
5. Apply the definitive model to achieve predictive analytics.
Cognitive Services in Power BI
Source: Microsot Azure Cognitive Services
With Cognitive Services in Power BI, you can apply various algorithms from Azure Cognitive Services to enrich your data. No Azure licence required
There are four cognitive services currently available for Power BI:
– Sentiment analysis. Azure Sentiment Analysis API identifies positive and negative opinions in forums, blogs, social networks, etc.
– Extraction of key phrases. Key Phrase Extraction API evaluates unstructured text and creates a list of key phrases.
– Language detection. Text Analytics identifies language, variants and even dialects.
– Image tagging. Computer Vision identifies people, animals, landscapes and objects in the image and returns the corresponding labels.
In order to implement Azure Cognitive ServicesAzure subscription is not required, but the Power BI Premium version is required.
Next-Step Consultants - Microsoft Partners & Power Platform Experts
Next-Step Consultores, we are Microsoft partner and experts in Power Platform: Power BI, Power APPs. If you would like more information, please contact us and we will be happy to advise you: