Ayesa helps Andalusia lead the way in AI for education

The company has been involved in developing Hipatia, a platform able to provide predictive insights by processing and analysing data.

The goal is to apply predictive analysis to the vast amount of data accumulated over the past 20 years on the Séneca system, something that will allow administrative and educational needs to be anticipated.

Ayesa, a global provider of technology and engineering services, is helping Andalusia lead the way in the use of AI to optimise educational management.

As such, it has recently developed Hipatia, an AI-powered platform able to offer users predictive insights based on the processing and analysis of data.

Created by Ayesa as part of a joint venture alongside Datrik Intelligence, and co-financed by the ERDF, it will allow the Regional Department for Education and Vocational Training to identify both administrative and educational needs before they become apparent.

The new solution has already been used to model various use cases. These include estimating numbers of teachers and students for various courses, predicting factors that contribute to truancy and optimising the learning process for vocational training.

In order to carry out predictive analysis, Hipatia draws on the vast amount of data accumulated on Séneca. This platform, launched in 2002, was also developed and implemented by Ayesa alongside the Regional Government of Andalusia. Its purpose is to manage schools, educational support services, inspections, curricula and other aspects of the region’s education system.

Séneca currently has academic information on more than 3.9 million students, records for 825 million grades and data on 711 million absences. By using data stored on this system, Hipatia is able to model the trajectory of students from primary school through to sixth form. This allows schools to optimise their planning and take the appropriate measures, such as hiring the correct numbers of teachers in each subject and taking steps to prevent students from dropping out of secondary school.

Francisco Silveira, Head of the Information, IT and Digital Systems Service at the Regional Department for Education and Vocational Training in Andalusia, explains: ‘Hipatia has allowed us to identify factors that contribute to students dropping out of school as well as variables that keep them engaged. Based on this, we have developed a tool to analyse school dropout’.

In terms of professional training, Hipatia has optimised the learning process and the issuing of qualifications. This includes modelling the allocation of places and subsequently providing learning support.

Ana Conde, Director of Employment and Innovation at Ayesa, explains how ‘the use cases put forward are applicable to any organisation or body involved in educational management, and the models developed could be implemented in other areas of Spain’.

As part of this project, Ayesa has provided data preprocessing services, and has also been involved in the training, evaluation and adjustment of the models. Furthermore, it has provided consulting services, which have seen it establish the hardware and software architecture needed to support the various predictive models developed. Training on the use of the predictive models has also been provided to staff at the Regional Department for Education and Vocational Training.

Among other technologies, Hipatia is based on the Hadoop Distributed File System (HDFS), Apache Spark, H2O Sparkling Water, the Parquet file format, Apache Zeppelin and BentoML.

Benefits of using AI in education

Hipatia is a clear example of how artificial intelligence can be harnessed to increase efficiency, responsiveness and productivity in education. Some examples of how it can be used include:

1. Automating the correction of exams and assessments: this saves teaching staff time and provides immediate feedback to students. It also allows patterns in performance and areas for improvement to be identified.
2. The analysis of educational data: this can help trends and patterns to be identified. These can then be used by teachers and schools to make informed decisions on teaching strategies, as well as detect difficulties in learning and personalise study programmes.
3. Providing online learning platforms: these can be used to provide personalised content and adjust learning routes based on the progress of students.
4. Simulating classes: these could allow skills to be gained in a controlled and safe environment.
5. Creating virtual assistants and chatbots to offer advice on courses and provide support to students.
6. Detecting plagiarism to prevent cheating.
7. Offering foreign language programmes that adapt lessons to the level of the student, provide feedback on pronunciation and personalise the learning experience.
8. Personalising content based on the individual needs of students: this involves adapting its difficulty, style and pace to maximise understanding and retention.
9. Creating virtual tutors that offer individual support to students without replacing the need for teaching staff: this includes providing explanations, answering questions and offering feedback in real time.