Abstract:
Water is polluting with different glass particles in Bangladesh. Most of the glass manufacturing
industries through their wastage in the river. Because of the indiscriminate release of some
industrial, domestic, and mining effluents into the environment and also on living species, water
pollution is now a global issue. Numerous strategies for the control, remediation, cleaning, and
purification of the water system from the source and at the end of the delivery line should have
been developed as a result of the negative impacts of water pollution on humans, animals, and
the ecosystem. Among all the methods used are membrane separation, biological precipitation,
adsorption, and photo catalysis. Besides, the development of water purification methods that are
less expensive, more cost-effective, and simpler to operate are crucial. Because they have the
potential to lessen the surface tension that exists between two immiscible liquids, surfactants and
bio surfactants are used in the processes of water treatment. Bio surfactants derived from natural
sources have gained attention due to their low cost, low impact on the environment, and unique
properties that make them useful in conjunction with nano materials to boost their activity and
performance. The use and performance of bio surfactant nanomaterial systems in water
purification processes are the subject of this review. Water samples from the water source near
some glass manufacturing companies and tested the water. With the report of these tested water,
a dataset of different glass particles is created. Then after this, using machine learning techniques
like Naive Bayes, KNN, and Random Forest algorithm different models are created and
compared their accuracy of predicting the water pollution caused by the glass particles. Among
them Naïve Bayes model performs with 92% accuracy whereas the KNN model performs with
93.6% accuracy and Random Forest stands out with 96% accuracy which is higher than the other
models.
