The invention discloses a simplified big data cleaning mode. The mode involves a big data cleaning task unit library, a Spark SQL module, a Spark-ETL SDK module, an assembly line configuration moduleand a webpage service platform composed of a webpage client and a webpage server. The mode is characterized in that a user adds a required cleaning unit and an algorithm task needed to be executed through the webpage server; the Spark SQL module receives the required cleaning unit and the algorithm task needed to be executed from the webpage server and realizes a data cleaning function; the cleaning unit and the algorithm task are extracted from the big data cleaning task unit library; and the assembly line configuration module submits an SDK-defined interface to a working server for performing effective operation. In the whole process, a definition file of a big data cleaning assembly line only needs to be written and configured, and no complex code programming is required; and especially, big data cleaning can be finished by utilizing Spark without a strong Spark programming ability.