All digitalized information is collected in a specialized database, where it is grouped by subject area. When a user enters a text query, it is also converted into a numeric vector. The system compares this vector with the database and finds the most relevant "piece" of data. This information is then added to the user's query and fed into the GPT language model. This way, the model gets not only the user's query, but also additional context, allowing it to provide more accurate and useful answers. The choice of this technology was due to its ability to efficiently process large volumes of data and provide relevant answers to user queries. With its help, it was possible to implement the following chatbot use cases.
Description of the MCN Telecom solution A smart chatbot can solve a wide range of tasks that make teachers' work easier: Come up with a lesson on a specific topic For example, create a lesson plan on spatial modeling armenia cell phone number list from a constructor, divide it into 6 stages and distribute the time so as to fit into 40 minutes. Manage your study load The chatbot offers the optimal distribution of educational material by week, taking into account the characteristics of the selected curriculum and the age of the students. Providing definitions and explanations The chatbot provides clear and understandable definitions or explains complex concepts based on the learning center materials. Answers to pedagogical questions Teachers can ask questions about teaching methods, and the chatbot provides them with verified answers from the knowledge base.
Here's what it looks like: Login The teacher enters the chatbot interface through the work portal. Entering a query In the text field, the teacher enters his request, for example: “Create a lesson plan on the topic ‘Spatial modeling’.” Processing request The chatbot analyzes the request, adds additional context from the knowledge base and passes it to the language model. Getting a response The chatbot returns to the teacher a detailed lesson plan, broken down into stages and scheduled by time. Results of implementation As a result of the pilot project, in which 48 teachers from 30 educational institutions participated over a period of 4 months, the following results were achieved.