Attention!
To participate in the forum of this paper and to receive the notifications, it is compulsory to activate it beforehand for your user. To do so, click on the button "subscribe". a little further down. You will be able to choose whether you want to receive all notifications in your email or only the replies to your comments.
Enjoy the Congress!
English (UK) 
Español
At the University of Guadalajara, tutoring is basic as a support to students, the tutor is a facilitator in their orientation within the institution, but unfortunately there are not enough staff to support them in that management, so the chatbot proposal can support in rudimentary tasks in certain procedures, to follow up students. The chatbot can have greater coverage and support in basic tasks.
The use of Artificial Intelligence and the fact that chatbots can be trained today contribute to facilitate repetitive tasks, for this case we used the Python programming language, JupyterNotebook, the pre-trained Llama model that facilitated the work, the continuous interaction and the guided messages and finally the institutional information letter.
Hello, congratulations for the presentation and the work, why did you apply Llama as a large scale language model, what advantages did you observe for applying Llama ? Greetings!
Hello,
Good morning. Thank you for the question. We use Llama because the model was trained from there before being passed to execution from a chatGPT environment. From Llama we ensure a desirable level of training (with two measurement parameters) to ensure that it will answer specific questions and then we can place it in a web environment.
Again congratulations on the project, another question, how to mitigate the ethical issues involved in the use of chatbots? Thank you!
Hello,
Good day. Thank you, very kind. Regarding your second query, which is very important by the way, we handled this aspect from the training part, although only a part of it. On the one hand, evaluations of the answers were carried out to assess the certainty and consistency or possible biases. On the other hand, we do not ask for the name of the person who consults so that this information is not saved, we indicate that it is an automated assistant not a substitute for the teaching tutor, the types of questions that can be asked, the data that were used to train it, as well as the possibility of indicating inadequate answers. In the next stage, a pilot test is expected to be carried out with several students to evaluate their perception and make any adjustments deemed appropriate and to detect patterns of misuse.
I hope I have answered your question.
Greetings!