Course Recommendation System based on Natural Language Processing
Smrithi Sumesh
M3 Signal Processing
Dept. of Electronics and Communication College of Engineering Trivandrum tve21ecsp15@cet.ac.in
Prof. Aswini S. H.
Assistant Professor
Dept. of Electronics and Communication College of Engineering Trivandrum aswinish@cet.ac.in
Abstract—Computers are actually capable of understand hu- man languages thanks to the era referred to as NLP (Natural Language Processing). Deep grammatical and semantic analysis frequently makes use of words as the primary unit. The principle goal of NLP is regularly word segmentation. This undertaking can be used to deal with the sensible problem of widespread structural variations between conventional and multi-modal envi- ronments and diverse statistics modalities. The challenge presents the basics of the multi-modal function extraction approach that uses deep gaining knowledge of. A advice device primarily based on content material is also contributed by the undertaking. Based on the person’s previous behaviour or express feedback, it recommends additional items which might be just like what they already like the usage of object functions. A course recommenda- tion gadget, in a nutshell, is a device that suggests the subsequent piece of content based on what got already seen and loved. Advice structures are utilized by services like Spotify, Netflix, and Youtube to signify the subsequent movie or track you need to watch primarily based on what you’ve got already watched or heard. Primarily based at the filtering records, the advice gadget foresees gadgets which can realise and provide excessive-potential content that has been chosen by the consumer. Primarily based at the customers’ various searches, a recommendation machine become evolved to indicate courses to them. Users now have an easier time finding the proper guides based on their searches thanks to this machine. To determine availability, the system employs the TF-IDF algorithm and content material-based totally filtering. Suggestions for customers who use the collaborative filtering technique had been researched in earlier studies.
Index Terms—Recommender System, Feature Extraction, Nat- ural Language Processing, Collaborative Filtering
