SELFIE Helper

SELFIE (Self-reflection on Effective Learning by Fostering the use of Innovative Educational technologies) is a free tool designed to assist schools in embedding digital technologies into teaching, learning and assessment.  However, in many cases, school representatives have questions regarding the registration process.

They have to ask SELFIE representatives in each involved country of the project and wait for their response. This approach does not allow them to finish the registration process before they get the answer and causes more work for SELFIE experts per country.

One of the key outputs of the SHERPA project is the SELFIE Helper, a component based on chatbot technology that is responsible for providing automated step-by-step assistance to schools when registering to the SELFIE platform.

The goal of SELFIE Helper is to facilitate schools in registering and getting started with SELFIE, so they can make use of SELFIE ‘s potential as a tool for self-evaluation of the school’s digital capacity.

To find out more, take a look at the SELFIE HELPER promo leaflet, available in English and Greek.

You can try out SELFIE Helper here.

The SELFIE Helper takes the form of a chatbot driven by Artificial Intelligence (AI) and Case-Based Reasoning.

Ultimately, the SELFIE Helper is expected to significantly boost wide-scale adoption of SELFIE across Europe and at the same time reduce the human effort needed to scaffold constantly expanding SELFIE usage. These gains should contribute significantly to the overall sustainability of the SELFIE ecosystem.

The SELFIE Helper system consists of 4 modules:

  • The Chatbot Interface
  • The CBR Inference Engine
  • The Knowledge Base
  • The Backend Management System

Queries posted by SELFIE school representatives via the Chatbot Interface are forwarded to the CBR Inference Engine which activates a diagnosis process capable of identifying any existing, appropriate similar case/answer stored in the Knowledge Base.

Here is the draft architecture foreseen for the SELFIE Helper:

SELFIE Helper architecture

The system can be proven useful and time saving for the SELFIE helpdesk service and the schools, considering that there is a strong likelihood of repeated problems emerging from initial use of SELFIE by schools.

Short Description of SELFIE Helper four modules

1. Chatbot Interface

Chatbot Interface is a web application that serves as a publicly available front-facing interface to interact with CBR Inference Engine application programming interface (API) and Knowledge Base service. CBR Inference Engine is used for submitting questions and getting a response, provided that an algorithm is able to find a suitable answer from available data stores.

The chatbot is available at http://helper.sherpa4selfie.eu/

2. CBR Inference Engine

The CBR Inference Engine, is the module which is capable of answer selection from a predefined closed set of Questions and Answers (Q&A) pairs, based on user’s questions. Its main core is based on the ChatterBot platform which is a machine-learning based conversational dialog engine generates responses based on collections of known conversations. It uses a selection of machine learning algorithms to produce different types of responses. The more input that Chatter-Bot receives increase the accuracy of its responses.

3. Knowledge Base

SELFIE Helper, includes a database that contains a set of questions and answers, described as Knowledge Base. The Knowledge Base consists of a set of questions and answers. Questions and answers are organized into categories or topics. Each category includes several questions and answers, while each question-answer corresponds to exactly one category or topic. The concept category or topic is considered equivalent.

Τhe SELFIE Helper KΒ provides multilanguage support for End Users. To address the questions/answers in the supported languages, with the aid of a back-end management system.

4. Backend Management System

The web-based Backend Management System (BMS) supports the Master Selfie Expert and the Country Selfie Experts categories of users. Overall, the BMS supports several roles, with some of those being explicit and some implicit.



The initial internal evaluation results revealed that SELFIE Helper

  • succeeds in guiding the SELFIE users through the registration process.
  • is easy and straightforward to use tool

Though more detailed evaluation results are expected after the official pilot with schools (of the SHEPRA project) that will make use of the SELFIE Helper.