Next Level Business Using Data Science / AI

Data science is an inter-disciplinary field that uses scientific methods, processes, algorithms and systems to extract knowledge and insights from many structural and unstructured data. Data science is related to data mining, deep learning and big data. In computer science, artificial intelligence (AI), sometimes called machine intelligence, is intelligence demonstrated by machines, not unlike the natural intelligence displayed by humans and animals. Leading AI textbooks define the field as the study of "intelligent agents": any device that perceives its environment and takes actions that maximize its chance of successfully achieving its goals. Colloquially, the term "artificial intelligence" is often used to describe machines (or computers) that mimic "cognitive" functions that humans associate with the human mind, such as "learning" and "problem solving".


  1. The front-end website or app service collects historical data of user-movie interactions, which are represented in a table of user, item, and numerical rating tuples.

  2. The collected historical data is stored in a blob storage.

  3. A DSVM is often used to experiment with or productize a Spark ALS recommender model. The ALS model is trained using a training dataset, which is produced from the overall dataset by applying the appropriate data splitting strategy. For example, the dataset can be split into sets randomly, chronologically, or stratified, depending on the business requirement. Similar to other machine learning tasks, a recommender is validated by using evaluation metrics.

  4. Azure Machine Learning is used for coordinating the experimentation, such as hyperparameter sweeping and model management.

  5. A trained model is preserved on Azure Cosmos DB, which can then be applied for recommending the top k movies for a given user.

  6. The model is then deployed onto a web or app service by using Azure Container Instances or Azure Kubernetes Service.


This scenario covers the back-end components of a web or mobile application. Data flows through the scenario as follows:

  1. The API layer is built using Azure Functions. These APIs enable the application to upload images and retrieve data from Cosmos DB.

  2. When an image is uploaded via an API call, it's stored in Blob storage.

  3. Adding new files to Blob storage triggers an Event Grid notification to be sent to an Azure Function.

  4. Azure Functions sends a link to the newly uploaded file to the Computer Vision API to analyze.

  5. Once the data has been returned from the Computer Vision API, Azure Functions makes an entry in Cosmos DB to persist the results of the analysis along with the image metadata.

These architectures are only meant to outline a basic Data Science/AI pipeline. We understand that your small business will require a custom solution to find the balance between cost and performance.


All it takes to get started is a conversation. Send us an email request.