A VEHICLE RECOMMENDATION SYSTEM FOR SALES

Abstract

The study explores the integration of the BERT4Rec model, a Natural Language Processing (NLP) model, into recommendation systems for the used automobile market. The model aims to improve user experience and sales by analyzing user preferences and past interactions. The study’s methodology includes feature encoding, data cleaning, sequence modeling prepara- tion, model training, and evaluation. The study aims to determine the effectiveness of the integrated BERT4Rec model in resolving the issue and enhancing recommendation accuracy within the used car market. Preliminary results show a 70% accuracy rate, indicating sig- nificant progress in recommendation systems. These findings could significantly impact the industry by improving user experiences and increasing sales in e-commerce platforms operating in the used automobile market.

‫ﺗﺴﺘﻜﺸﻒ‬‫ﺍﻟﺪﺭﺍﺳﺔ‬ ‫ﺩﻣﺞ‬ ‫ﻧﻤﻮﺫﺝ‬ ‫‪BERT4Rec‬‬ ‫‪،‬‬ ‫ﻭﻫﻮ‬ ‫ﻧﻤﻮﺫﺝ‬ ‫ﻣﻌﺎﻟﺠﺔ‬ ‫ﺍﻟﻠﻐﺎﺕ‬ ‫ﻟﻄﺒﻴﻌﻴﺔ‬ ‫‪NLP‬‬‫‪،‬‬ ‫ﻓﻲ‬ ‫ﺃﻧﻈﻤﺔ‬ ‫ﺍﻟﺘﻮﺻﻴﺔ‬ ‫ﻟﺴﻮﻕ‬ ‫ﺍﻟﺴﻴﺎﺭﺍﺕ‬ ‫ﺍﻟﻤﺴﺘﻌﻤﻠﺔ‪.‬‬ ‫ﻳﻬﺪﻑ‬ ‫ﺍﻟﻨﻤﻮﺫﺝ‬ ‫ﺇﻟﻰ‬ ‫ﺗﺤﺴﻴﻦ‬ ‫ﺗﺠﺮﺑﺔ‬‫ﺍﻟﻤﺴﺘﺨﺪﻡ‬ ‫ﻭﺍﻟﻤﺒﻴﻌﺎﺕ‬ ‫ﻣﻦ‬ ‫ﺧﻼﻝ‬ ‫ﺗﺤﻠﻴﻞ‬ ‫ﺗﻔﻀﻴﻼﺕ‬ ‫ﺍﻟﻤﺴﺘﺨﺪﻡ‬ ‫ﻭﺍﻟﺘﻔﺎﻋﻼﺕ‬ ‫ﺍﻟﺴﺎﺑﻘﺔ‪.‬‬ ‫ﺗﺘﻀﻤﻦ‬‫ﻣﻨﻬﺠﻴﺔ‬ ‫ﺍﻟﺪﺭﺍﺳﺔ‬ ‫ﺗﺮﻣﻴﺰ‬ ‫ﺍﻟﻤﻴﺰﺍﺕ‬ ‫ﻭﺗﻨﻈﻴﻒ‬ ‫ﺍﻟﺒﻴﺎﻧﺎﺕ‪،‬‬ ‫ﻭﺇﻋﺪﺍﺩ‬ ‫ﻧﻤﺬﺟﺔ‬ ‫ﺍﻟﺘﺴﻠﺴﻞ‪،‬‬ ‫ﺗﻬﺪﻑ‬‫ﺍﻟﺪﺭﺍﺳﺔ‬ ‫ﺇﻟﻰ‬ ‫ﺗﺤﺪﻳﺪ‬ ‫ﻣﺪﻯ‬ ‫ﻓﻌﺎﻟﻴﺔ‬ ‫ﻧﻤﻮﺫﺝ‬ ‫ﻭﺍﻟﺘﺪﺭﻳﺐ‬ ‫ﻋﻠﻰ‬ ‫ﺍﻟﻨﻤﺎﺫﺝ‪،‬‬ ‫ﻭﺍﻟﺘﻘﻴﻴﻢ‪.‬‬ ‫‪BERT4Rec‬‬‫ﺍﻟﻤﺘﻜﺎﻣﻞ‬ ‫ﻓﻲ‬ ‫ﺣﻞ‬ ‫ﺍﻟﻤﺸﻜﻠﺔ‬ ‫ﻭﺗﻌﺰﻳﺰ‬ ‫ﺩﻗﺔ‬ ‫ﺍﻟﺘﻮﺻﻴﺎﺕ‬ ‫ﺩﺍﺧﻞ‬ ‫ﺳﻮﻕ‬ ‫ﺍﻟﺴﻴﺎﺭﺍﺕ‬ ‫ﺍﻟﻤﺴﺘﻌﻤﻠﺔ‪.‬‬‫ﻭﺗﻈﻬﺮ‬ ‫ﺍﻟﻨﺘﺎﺋﺞ‬ ‫ﺍﻷﻭﻟﻴﺔ‬ ‫ﻧﺴﺒﺔ‬ ‫ﺩﻗﺔ‬ ‫ﺗﺒﻠﻎ‬ ‫‪70%‬‬ ‫‪،‬‬ ‫ﻣﻤﺎ‬ ‫ﻳﺸﻴﺮ‬ ‫ﺇﻟﻰ‬ ‫ﺗﻘﺪﻡ‬ ‫ﻛﺒﻴﺮ‬ ‫ﻓﻲ‬‫ﺃﻧﻈﻤﺔ‬ ‫ﺍﻟﺘﻮﺻﻴﺔ‪.‬‬ ‫ﻳﻤﻜﻦ‬ ‫ﺃﻥ‬ ‫ﺗﺆﺛﺮ‬ ‫ﻫﺬﻩ‬ ‫ﺍﻟﻨﺘﺎﺋﺞ‬ ‫ﺑﺸﻜﻞ‬ ‫ﻛﺒﻴﺮ‬ ‫ﻋﻠﻰ‬ ‫ﺍﻟﺼﻨﺎﻋﺔ‬ ‫ﻣﻦ‬ ‫ﺧﻼﻝ‬ ‫ﺗﺤﺴﻴﻦ‬‫ﺗﺠﺎﺭﺏ‬ ‫ﺍﻟﻤﺴﺘﺨﺪﻡ‬ ‫ﻭﺯﻳﺎﺩﺓ‬ ‫ﺍﻟﻤﺒﻴﻌﺎﺕ‬ ‫ﻓﻲ‬ ‫ﻣﻨﺼﺎﺕ‬ ‫ﺍﻟﺘﺠﺎﺭﺓ‬ ‫ﺍﻹﻟﻜﺘﺮﻭﻧﻴﺔ‬ ‫ﺍﻟﻌﺎﻣﻠﺔ‬ ‫ﻓﻲ‬ ‫ﺳﻮﻕ‬‫ﺍﻟﺴﻴﺎﺭﺍﺕ‬ ‫ﺍﻟﻤﺴﺘﻌﻤﻠﺔ‪.‬‬