Unless you’ve been living under a rock, you won’t be surprised to know that the main theme of CloudWorld this year at Las Vegas was generative AI.

Almost every Oracle product from databases, to Fusion applications, to OCI, APEX and Analytics to name but a few announced generative AI functionality at the recently concluded Oracle CloudWorld 2023 in Las Vegas that I attended.

My journey from the U.K. to Las Vegas was "eventful" and could be a separate blog in its own right!

​Another point made that I was admittedly unaware of is that Oracle Analytics Server and Oracle Analytics Cloud only share a very small amount of source code. This is the main reason why there is a significant difference in the Oracle Analytics Cloud and Oracle Analytics Server release cycles; every 2 months for Oracle Analytics Cloud and annually for Oracle Analytics Server.

​In the integrated data preparation area, PII (Personally Identifiable Information) is coming soon. My take on this is that this will be another recommendation on the data preparation step. The web based semantic modeller support for EPM & Essbase data sources is also something eagerly awaited and, on the roadmap, marked as “later”.

The information from this strategy update is publicly available and can be found here.

Catalog Manager
​
It was recently announced that the Catalog Manager GUI will be desupported from the January 2024 Oracle Analytics Cloud update. The command line interface will still be available to use from that point. The January 2024 update will not ship with the Catalog Manager client tool however, you’ll still be able to use existing installations of the Catalog Manager client if needed. The strategic plan is for the Catalog Manager client tool to be completely replaced by a web-based Catalog Manager. There are also OAC catalog APIs in the works as well.
 
T.K. Anand had a session which I couldn’t attend that provided some more context and information about Larry Ellison’s announcement about Oracle Cloud Data Intelligence Platform. I think that this is going to bring Oracle Analytics and the Fusion Analytics platform under one umbrella in conjunction with generative AI. If this is the case, then it will simplify the Oracle Analytics offering doing away with all the confusion around which of the products are best suited to specific customers. I’ll be keeping an eye on developments on this.
 
My PresentationI presented at CloudWorld for the first time, and it was an interesting experience. Tim German, the Version 1 US EPM Practice Head was a co-presenter. The topic we covered was “from data to insights with Oracle Analytics”. We had a really great turnout with standing room only and about 100 people in attendance, we had to cram a lot of content into a 20-minute theatre session to demonstrate the capability of Oracle Analytics to enable users to gain insights from their data assets. The key takeaways of the session were that Oracle Analytics:
​
•       allows users to connect to a variety of data sources.
•       users can easily visualise and analyse data.
•       built-in machine learning and artificial intelligence capabilities
•       tools for data preparation.
•       collaboration and sharing features.

CloudWorld has been a positive experience, and I had the opportunity to meet a lot of people in person for the first time that I’ve only seen on Zoom and Teams calls! As an Oracle ACE Pro, I enjoyed the Oracle ACE events which were good fun where there were over 100 Oracle ACEs in attendance at the ACE dinner! I also got to meet some of the Version1 Oracle ACEs in person too.

​I just hope that my luggage won’t be overweight on my way home with all the swag that I picked up!

In the dynamic landscape of data analytics, businesses are in search of a comprehensive solution that seamlessly integrates data governance, self-service analytics, augmented insights, and robust reporting. Oracle Analytics emerges as a unified platform that stands out amidst the competition. Let's draw a comparison with other leading tools – Tableau, Power BI, and Qlik – and delve into Oracle's key differentiator: the Semantic Model, and how it empowers cross-platform compatibility.

Comparing Solutions:

In a landscape where specialised tools often excel in specific areas; some examples stand out:
​

• Tableau (Self-Service Analytics): Tableau offers user-friendly self-service analytics, allowing users to create captivating visualisations. However, it lacks the depth of governed analytics and comprehensive AI-powered insights.
• Power BI (AI-driven Insights): Power BI shines with its AI-driven insights and automation of data analysis. However, its capabilities can sometimes be limited when integrating insights with external reporting tools.
• Qlik (Data Governance): Qlik emphasises data governance and provides strong analytics capabilities. But it lacks the seamless integration required for a comprehensive analytics ecosystem.

In contrast, Oracle Analytics emerges as the bridge that seamlessly connects and combines these functionalities, underpinned by the powerful and unique Semantic Model. Unlike other vendor solutions that focus on one specific area of Analytics, Oracle Analytics unites governed analytics, self-service analytics, and augmented analytics within a single, cohesive platform.
​

• Business Analysts - Oracle Analytics provides self-service data visualisation and discovery tools for business analysts to easily access, prepare, analyse, and visualise data to gain insights. Features like data preparation, auto insights, smart visualisations and natural language processing enable faster and easier analysis.
• Data Scientists - For more advanced analytics, data scientists can use Oracle Analytics to build and compare machine learning models, perform clustering and predictions, and leverage R and Python scripts for advanced analysis. Oracle Analytics includes auto ML to make model building faster.

• Developers - Developers can embed analytics into applications using Oracle Analytics Cloud SDKs and REST APIs. They can build custom visualisations, dashboards, and data flows to extend Oracle Analytics capabilities. The Semantic Modeller incorporates data governance capabilities like business-friendly data sets, stewardship, master data management etc. Developers can build on these capabilities for enterprise data governance needs.
• Casual Business Users - With interactive dashboards, natural language query and conversational analytics, casual business users can intuitively explore data and find insights through guided analysis and AI assistant. Smart visualisation recommendations make it easy for novice users.
• Executives/Management - Interactive dashboards provide executives with ability to visualise and track KPIs and metrics to monitor performance. Natural language makes it easy to ask questions of data. Data storytelling features create narrated stories to simplify sharing insights.

The goal of Oracle Analytics is to empower all personas across organisations to make data-driven decisions through easy, intelligent, and connected analytics experiences. The versatility caters to both highly technical and business casual users.

A Path Forward: Embrace Unity with Oracle Analytics:
​

In the July 2023 update, Oracle Analytics Cloud has introduced a new feature called AutoML. Oracle AutoML, or Oracle Automated Machine Learning, is a feature within the Oracle Cloud Infrastructure (OCI) platform that has been accessible within OCI in the Autonomous Database. Its main objective is to automate the process of developing and deploying machine learning models. AutoML is designed to simplify and accelerate the model building process, particularly for users who may not have extensive expertise in data science or machine learning.
 
What exactly is AutoML?
​
Oracle AutoML provides a user-friendly interface that allows users to upload their data, specify the target variable they want to predict or classify, and set various model configuration options. The platform then automatically explores different algorithms, feature transformations, and hyperparameter settings to find the best-performing model for the given task.

​The AutoML process involves several steps, such as data preparation, feature engineering, model selection, hyperparameter tuning, and model evaluation. Oracle AutoML takes care of these steps behind the scenes, leveraging advanced techniques and computational resources to optimise the model building process.
Once the best model is identified, Oracle AutoML enables users to deploy the model for real-world use cases. The deployed model can be integrated into applications or used for making predictions and decisions based on new data.
Overall, Oracle AutoML aims to democratize the use of machine learning by automating complex tasks and reducing the need for specialized knowledge. It enables users to leverage the power of machine learning without the extensive time and effort typically required for model development, allowing them to focus more on deriving insights and value from their data.
AutoML in Oracle Analytics CloudAutoML has now been embedded in Oracle Analytics in the July 2023 update enabling users with no prior Machine Learning experience to use the right models with the correct hyperparameter settings.

Right clicking on the model and selecting the Inspect menu option gives you more information about the ML model. ​

The Oracle Analytics REST API Connector is now available in the September 2022 Oracle Analytics update in preview mode.

The connector now allows access to data directly from REST API endpoints.

​To access the REST API Connector, you will need to enable the feature in the Console System Settings preview section by switching on the toggle.

You will need to log off and back on again to apply the change to your instance. You should now see the REST API connector in the create > connections menu on the Oracle Analytics homepage.

In the next screen, you enter details of your REST API that you want to access. The basic information required for connecting are:
​

1. Base URL of the REST API
2. REST API endpoints
3. Authentication type
4. Authentication credentials

​
You can either enter the information in here directly or you can use a JSON template to enter all of the REST API information. A sample JSON template file can be found at the bottom of this post.

If you decide to enter information of the REST API for your connector using a JSON file, you will need the contents to match the syntax below. If your API has several endpoints that you will be using then enter each endpoint URL in the endpoints section.

​Ensure that the correct authentication type is entered. Once all completed, the connection can be used as a standard connection to create a dataset. .

During the testing of this new functionality, the REST API used BasicAuth authentication and the API key was supposed to be entered in the username field. The issue here is that the password is a mandatory field and can't be left blank. The way round this was to re-enter the API key in the password field as well.

Once the connection has been created, you can create a dataset using the REST API connection in the normal fashion.

In the connection panel of the New Dataset window expand the REST API connection to the AUTOREST schema to access the result sets of the endpoints. The RESULTS tables contain the data from the endpoints and the others contain metadata of the endpoints.

The new REST API connector is a great addition to the Oracle Analytics capability. It now gives users another means of bringing additional data in to uncover even more insights from your enterprise data assets.

Photo by Isaac Smith on Unsplash

The Oracle Analytics team have been busy working on the latest version of Oracle Analytics Cloud - 5.9 which is packed with a number of interesting features and enhancements that we'll cover in some detail in this blog post. 

There is now a permanent Search field and smaller Data, Visualisations and Analytics icons which have been relocated to the top of the side bar. This results in the reduction of wasted space on the left of the side bar.    

Progress Bars

The animated progress bars that are displayed when visualisations and data are loaded have been updated.

The thin blue horizontal progress bars that were visible in previous Oracle Analytics versions across the screen when a. visualisation or canvas was refreshed have been changed as you can see in the images above. 

TRIM Function
​
There is now one click access to TRIM whitespace from a text column in the Data Preparation stage of Data Visualisation. This removes whitespace from both sides of the selected column  

US ZIP Codes

​This new feature is available in a column that contains ZIP code data. Oracle Analytics suggests a recommendation step to repair any of the ZIP codes that have missing leading zeros.

Filter Enhancements

You can now move a Visualisation level filter to the Canvas level by simply dragging it from the side bar to the Canvas level filter section. You can also do the opposite; moving a filter from the Canvas level to a specific Visualisation.

Sorting Enhancements

​There is now the capability in OAC 5.9 to sort by multiple dimensional attributes. In the screenshot below, you can see that the state_name and county_name have sorts defined.

​Columns added to the Size and Tooltip of a visualisation can also be included in the visualisation sorting. 

You can also sort by a measure but note that this will overwrite any dimensional attribute sorting that has been previously defined for the visualisation. 

Export Limits

For those who still export data dumps to Excel ?, row limits have been increased. Formatted exports that include XLSX, PPTX, HTML and PDF files have been increased to 400,000 rows with instances that have 16-52 OCPUs. Instances with 2-12 OCPUs will have their limit increased to 200,000 rows. 

The limits of unformatted report exports (XML, Tab Delimited and CSV) have been increased to 2,000,000 rows for instances with 2-12 OCPUs. Instances with 16-52 OCPUs will see their limits rise to 4,000,000 rows. ​

There were previously 2 Area Visualisations in versions of Oracle Analytics prior to version 5.9. There are now 3 types; Area, Stacked Area and 100% Area. 

I personally think that the new Stacked Area visualisation is more "visual" and users can gain a quicker insight visually. 

These new map backgrounds give users more flexibility with their map visualisations. 

It can be useful to gain some insights into the Machine Learning models that Oracle Analytics uses. There have been some enhancements added to the Oracle Machine Learning (OML) inspect tab which exposes much more metadata about the Machine Learning model. 

There has been some renaming and reorganising of the tabs available in the Machine Learning Inspect window. The Details tab contains basic information about the Oracle Machine Learning model including input and output columns. 

The Related tab now shows in-depth information of the Oracle Machine Learning model which is stored in database metadata views. 

You can even access these views and build out visualisations to gain further insights and understanding of the Machine Learning models that you are using in your Analytics.

​The Oracle Analytics team have put together a YouTube video about it which you can view below. 

In previous versions of Oracle Analytics, Text Tokenization was indirectly available. OML ADW capabilities could be used. Francesco Tisiot recently blogged about this with a good example demonstrating a typical use case for textual analytics. 

In Oracle Analytics Cloud 5.9, Text Tokenization is natively available as a data flow step thereby taking away the need to access the ADW instance in order to create a context index on the text column to be analysed and then to link the tokens to the text field in your data that is being analysed. This is now done for you automatically.  

The Text Tokenization functionality only works with Oracle sourced datasets - ATP, ADW and Oracle On-Premise Databases. It makes use of the Oracle Text database functionality and the Database Analytics data flow step is only available for these Oracle sourced datasets. 

​In order to avoid this error:

Ensure that your user account that you use for the connection to the database has the correct privileges:

As you will have gathered, this 5.9 release of Oracle Analytics Cloud comes packed with a wide range of features and enhancements. The release is typically staggered whilst the upgrade is applied to the Oracle data centres in the various regions so watch this space! If you want to see further details of all of these new features, check out the Oracle's "What's New" document here.

If you’re intrigued with the features that Oracle plan to incorporate into the Oracle Analytics platform here is some information on an Oracle Analytics resource worth mentioning; the public roadmap which contains information on Oracle Analytics features that may be included in the platform with an idea of when these may become generally available.

A very quick blog post to provide some information about an issue that may arise with Data Visualisation projects based on manual SQL queries in Oracle Analytics Server 5.5 and above. These queries would have worked in previous versions of Oracle Analytics Server.

Essentially, manual multi line SQL queries throw an exception:
​

Oracle have confirmed that this is a defect that has a workaround. You will only encounter this issue if the SQL query test consists of several lines ending with carriage returns. 

​The workaround is to place a space character at the end of each line of the query text just before each carriage return. You can find more information about this here.

Version 5.7.2 of Oracle Analytics Cloud recently started rolling out to Cloud Customers. It comes with a couple of cool features which we'll go through in some detail. A full list of all the features can be found in the 'What's New' section of the Oracle Analytics Cloud documentation.

Performance Tile Enhancements

There is an update to the Performance Tile visualisation now that allows you to show a breakdown of the measure displayed by a dimensional attribute. There are options to show the dimensional attribute as a line chart, bar chart or as an area chart.

Database OML Models

Oracle Analytics Cloud 5.7 now gives users access to register and apply Machine Learning models from an Oracle database (On-Premise or Autonomous Data Warehouse). The main benefit of this feature is that there is no longer any need to physically move data around whilst training and running the model.

The other benefit of this approach is that the computing power of the host database of the Machine Learning Model is used when it is executed in Oracle analytics Cloud.

​

​

Once the model has been registered you can look at the properties of the model

Now that the Machine Learning model has been registered, to make use of it in Oracle Analytics Cloud, you create a Data Flow bearing in mind that your data set must be located in the same database as your Machine Learning model (the connection must be the same for the data set and the model).

The final step in the Data Flow specifies where the saves the data which is returned from applying the model to the dataset with the predictions which have been calculated. 

​A table needs to be defined that stores this data in the same database as the Machine Learning model and the data.

List Box Visualisation

Now this is a feature akin to the OBIEE dashboard prompt although this new visualisation provides a bit more in terms of functionality. You now have the ability to filter visualisations on the canvas at a more granular level. The List Box visualisation filtering can be set to apply to visualisations that you specify. which gives you much more control over how end users can interact with the data on the canvases.

​There is also the possibility to cascade filtering if you have multiple list box visualisations on a canvas in your project. This means that you can set up the filtering of one to filter on the other based on a selection.

There are some other features that are also included in this release which may be covered in a future blog post.

Photo by Mitchell Hollander on Unsplash

The Oracle Analytics product teams have been working round the clock to put the finishing touches to the latest version of Oracle Analytics Cloud which has just been released. Oracle Analytics Cloud 5.6 comes packed full of new features that aim to make your work of gaining insights into your data that much easier.

In this blog post, we will look at some of the new features that are now available in Oracle Analytics Cloud 5.6.

Maps with custom backgrounds

One feature that could change the way you visually interact with your data is the new method now available which now gives you the ability to use any image in a visualisation as a map layer. You can follow some easy steps to link the custom map layer to your data.

Upload an image in the Console > Maps > Image Backgrounds. The next step is to create a Map Layer and in this step you define the shapes (polygons, rectangles or circles) that you want to "map" to your data.

The next step is then to assign the map layer to a column in a Data Set editor (see below).

When you use this column in a visualisation, the default visualisation will be a map visualisation.

Dataset Editor Enhancements

There have been a few changes made to the Oracle Analytics interface that is used to make modifications to Datasets.

Some of the changes made to the Dataset Editor include:

Columns can now be deleted in the preview screen with a single click.

The properties of a column can now be viewed in a properties pane. You select a column from the data preview pane to access the column's properties.

You can apply a filter to your data set within the data set editor.

Regular expressions (Regex) can be used to search and replace patterns. This is done via the Replace feature in the dropdown of each column in the data set editor.

Custom calculations have a new option to duplicate the calculation which is useful in a scenario where you want to create a new calculation similar to one already created rather than having to start completely from scratch.

There is also a ‘Copy Expression to Clipboard’ option now available in the dropdown which I'm still trying to work out a use case for this functionality. Pasting into a text editor shows the expression:

Data Visualisation Enhancements

Columns and rows in pivot table and table visualisations can now be resized by dragging on the field left, right, up or down as required. I wasn't able to resize the rows in a table visualisation when I tested out this functionality.

The data preview section has been redesigned to enable more information to be displayed on screen. This has also been implemented in the prepare tab as well.

You can find more information on these enhancements and all other features that are included in the 5.6 release here.

Oracle Analytics Server 5.5.0 was recently released and I've been testing out the OAS installation process. This short blog post just goes through some issues that I encountered during this process related to access to Data Visualisation.

I did an out-of-place upgrade which went remarkably smoothly and imported a BAR file from a 12.2.1.4 OBIEE instance into the new OAS instance. The problem I had was to do with access to Data Visualisation.

When I tried to access the Data Visualisation page (server:port/dv/ui), I got an insufficient permissions error:

After some digging around, it transpires that there were some WebLogic Application policies that weren't assigned to any Application Roles. Oracle Support pointed me in the direction of this Support Document - Oracle Support Document 2644363.1. The 12.2.1.4 BAR file import wrote over some of the default WebLogic application permissions.The actio.admin permission needs to be granted to the BI Service Administrator application role. This is all done in em (WebLogic Enterprise Manager).

To compartmentalise the Data Visualisation security, I'd highly recommend creating 2 Application Roles:

1. DV User
2. DV Administrator

These roles will have the appropriate Application Policies and Resource Types assigned. These roles can then be assigned as required to the appropriate users.For example, the DV Administrator role can be assigned to a WebLogic administrator or to the BI Service Administrator role.

I was about to publish this blog and the Oracle Support folk just put together a new Support Document based on this issue with overwritten WebLogic application policies after an import of a 12.2.1.4 BAR file into an OAS instance..Hope this saves someone a bit of time.

The Oracle Analytics Product Management team descended on London ahead of the Oracle OpenWorld Europe event for the Oracle Analytics Summit London event held at The Soho Hotel in London on the 11th of February earlier this week.

The day kicked off with a Partner Advisory Council session from the Product Management team which included a look ahead at the roadmap for Oracle Analytics as well as a Q&A session which was very informative.

There were several other sessions throughout the day providing participants with a wealth of knowledge on all things Oracle Analytics.

The biggest announcement of the day was undoubtedly Oracle being named a visionary in Gartner's Magic Quadrant for Analytics and BI Platforms. The news was hot off the press and received mixed reactions.

Oracle has had a torrid time in the eyes of the Market Researchers in recent times with the likes of Gartner completely excluding Oracle from the Analytics Magic Quadrant not too long ago. Gartner's justification was regarding the fact that Oracle's Analytics offerings provided little or no self service features which is an opportunity that Oracle missed with the likes of Qlik, Tableau & Power Bi to name but a few that filled the gap in the Self Service Analytics space.

The Oracle Analytics team has filled the Self Service void with a variety of features and products including products and features like Oracle Analytics Desktop, Data Visualisation, Natural Language Processing and all the Augmented capabilities that enable end users to get to their insights with little or no IT intervention. This focus and attention led to Gartner bringing Oracle back into the Magic Quadrant as a niche player.

Oracle has made huge strides in the self service analytics space at the expense of the governed analytics capabilities that hasn't seen much development and enhancements in recent times.

The governed analytics part of the Oracle Analytics product is very mature and is its unique selling point. You hear of many stories where end users have acquired a self service analytics tool and plug it into OBIEE's semantic model. Most of the "new age" analytics tools are geared around self service and there still appears to be a huge demand for governed analytics which Oracle Analytics provides alongside its self service capabilities.

Some have said that Oracle may have taken their eye off the ball in order to focus all attention on getting back into Gartner's good books. As mentioned previously, a lot of attention has been focused on self service capabilities possibly to the detriment of governed analytics capabilities.

There was a mix of the Oracle Analytics Product Management team, Partners and Customers in attendance at the Oracle Analytics Summit and it was great to see and hear first hand from the Product Management team.