A Hybrid Travel Recommender Model Based on Deep Level Autoencoder And Machine Learning Algorithms

Abstract

This research investigates the application of autoencoders in processing travelogues written in the Malayalam language on Facebook. The main objective is to harness the capabilities of autoencoders to learn a compressed representation of the input data and employ it to train various machine learning models for enhanced accuracy and efficiency. The major challenge of unavailability of a benchmark dataset in the Malayalam language for the travel domain was overcome by employing NLP techniques on the unstructured, lengthy, imbalanced travelogues, applying some additional filtering methods, and the creation of an exclusive Part of Travel Tagger (POT Tagger) along with lookup dictionaries. As this pioneering work focuses on Malayalam travel reviews posted on social media, the model presents a valuable opportunity for extension to other low-resourced Indian languages. The study follows a two-step approach. Initially, an autoencoder neural network architecture is utilized to encode the travelogues into a lower-dimensional latent space representation. The encoder network adeptly captures crucial features and patterns within the data. The compressed representation obtained from the encoder is then fed into the decoder, which reconstructs the original travelogues. Subsequently, the encoded model is employed to train diverse machine learning models, including logistic regression, decision tree classifier, support vector machine (SVM), random forest classifier (RFC), K-nearest neighbours (KNN), stochastic gradient descent (SGD), and multilayer perceptron (MLP). By utilizing the encoded features as inputs, these models effectively learn from the concise representation of the Malayalam travelogues. Experimental results reveal that the trained machine learning models, using the encoded features, achieve higher accuracy rates compared to conventional approaches. This improvement demonstrates the effectiveness of autoencoders in capturing and representing vital characteristics of the Malayalam travelogues on Facebook. By leveraging capabilities of autoencoder model, we successfully learned a compressed representation of the input data, attaining an impressive validation accuracy of 95.84%. This finding highlights the potential of autoencoders to enhance the overall accuracy and efficiency of travel recommendation systems for Malayalam users on social media platforms.