Artificial intelligence

Artificial intelligence (AI) is the simulation of human-like intelligence on a machine. It can vary vastly in type and can be used to solve many problems, such as finding an optimal solution to a complex problem or finding patterns in large amounts of data. AI involves creating software that is capable of learning, reasoning and problem-solving. The field is constantly evolving, and it has the power to increase efficiency, improve decision making and create innovative solutions to complex problems.

Machine learning

A large portion of the field of AI is Machine learning (ML). It is an extensive topic and can be used in a wide variety of applications. ML is when a machine learns by itself from large amounts of structured data. Based on how the data is presented for the ML model, it can learn different abilities. For example, imagine a media website containing millions of images with accompanying tags that briefly describe the image. With this tag data, an ML model can learn to predict what tags a user would put on a new untagged image. ML models are especially great at extracting patterns in large amounts of data.

When a web application is up and running, it commonly generates large amounts of data over time, especially when it is built with Drupal. This data accumulates continuously but is rarely exploited to improve the quality of the product. If, for example, we take the previously mentioned media website, the tag prediction could be implemented in such a way that if a new user uploads a new image, the image tags could automatically get predicted and presented as suggestions for the user that uploaded the image. The media website is just one workflow example and can be extended to many others. ML can also be used to find anomalies in data and group data by similarity, meaning new patterns and trends can be discovered from existing data. 

Recent progress in machine learning has allowed machine learning to understand natural language, meaning that more text-related understanding and text generation are possible. 

Implementing AI

Here at Websystem, we combine the power of Drupal with artificial intelligence to produce tailored solutions to fit the data available and the product required. We offer guidance in using the data more effectively and give new ideas on what we can accomplish with the data available. We can also contribute to building data collection systems to collect data for future usage of AI.

Areas that are especially useful to implement AI within are:

  • Recommendation systems
  • Computer vision
  • Natural language processing
  • Data Science

And some examples of how we implement and use AI:

  • Define API:s to interface with machine learning models
  • Machine learning models embedded in the backend
  • Data collection systems to improve data quality and collect new data
  • Create new ML models to fit the required use case
  • Solve computationally complex problems with AI