An adaptive learning system aims to provide learning that is adapted to a learner's current status, different from the traditional classroom experience. A key element of an adaptive learning system is the recommendation system, which provides the most suitable resources based on learner profiles. Recommending the most appropriate learning resources to learners has always been a challenge in the field of e-learning. Thus, learners may have difficulties in choosing the appropriate material when faced with a large volume of recommended material during their learning process. This challenge led us to implement a chatbot to help learners improve their learning experience and knowledge. New solutions use artificial intelligence (AI) techniques such as machine learning (ML) and natural language processing (NLP). The use of our chatbot integrated in Moodle, named LearningPartnerBot, provides learners a personalized recommendation of learning objects according to two strategies, one based on their knowledge level (KL) and the other based on their learning style (LS). The objective of this article is to compare the learning outcomes obtained after the realization of the two experiments based on these two approaches centered mainly on the use of the LearningPartnerBot. Consequently, the approach of recommending learning objects based on the knowledge level gave promising results by guaranteeing a more adapted learning to the learners.