Standardize data from multiple sources

Company

Our client operates in the used vehicle resale sector. A major player in the industry, the company uses a web-based platform to monitor and evaluate vehicles on the market.

Challenge

Standardize the information presented and the presentation format of available vehicles to facilitate decision-making.

• Considering that there are between 3 and 4 million vehicles on the market every year, monitoring all available vehicles and identifying the right opportunities represents a challenge in itself.

• The vehicles on the platform come from multiple sources. As each supplier has its own vehicle classification and identification system, the information received by our customer is not homogeneous and differs in terms of nomenclature, description elements and presentation format.

Link to customer data :

• In our customer's case, an additional level of complexity is added, as they have developed their own criteria and vocabulary for evaluating the potential of the vehicles available, so that the data is comparable...

Solutions

• With the help of models based on natural language recognition (NLP) and Large Language Model (LLM) such as GPT3.5, the tools developed by Explorai aimed to :

• Collate information on available vehicles from several different sources

• Standardize the data for each vehicle and associate the attributes with those used internally by the customer to assess their market potential.

Results

• A better customer experience thanks to a standardized presentation of data from a variety of sources, making it easier to compare vehicles, regardless of their origin.

• More informed decision-making thanks to data presented according to user needs (attributes, vocabulary and classification)