Otonomo Technologies has been granted a patent for a method and system that normalizes data and data formats of automotive data obtained from various sources. The system includes a data collector, a data manipulating platform, and a computer processor that executes manipulating modules to normalize the data entries. Machine learning algorithms are applied to update the normalization rules based on the diversity of the data entries. The normalized data entries can be used as a common data language for automotive data consumer software applications. GlobalData’s report on Otonomo Technologies gives a 360-degree view of the company including its patenting strategy. Buy the report here.
According to GlobalData’s company profile on Otonomo Technologies, Vehicle telematics based risk analysis was a key innovation area identified from patents. Otonomo Technologies's grant share as of September 2023 was 52%. Grant share is based on the ratio of number of grants to total number of patents.
Patent granted for normalizing automotive data from multiple sources
A recently granted patent (Publication Number: US11748370B2) describes a method and system for normalizing data and data format of automotive data associated with connected vehicles. The method involves obtaining data entries from various sources, which are presented in different data formats. The system enables the selection and ordering of manipulating modules that can manipulate the data or data format of the data entries. These selected and ordered manipulating modules are then executed using a computer processor to normalize the data entries according to a predefined data format.
To improve the normalization rules, the method applies machine learning algorithms to the data entries. This involves classifying the various data types, formats, names, usage, origin, and content, in order to learn and model the format diversity. The model derived from the machine learning process is then used to enhance the normalization rules based on the actual diversity of the data entries. The machine learning algorithms are specifically applied to sensor records to classify and identify metadata and sensor types, define policies per sensor type, and apply these policies to data records belonging to the same sensor type.
The manipulating modules mentioned in the claims are software modules that contain instructions in a computer-readable medium. These modules can manipulate the data type, data name, data format, and data content of the data entries according to the normalization rules. The resulting normalized data entries, which conform to the predefined data format, can be used as a uniform or common data language to support various use cases for automotive data consumer software applications.
The patent also describes a system for implementing the method. The system includes a data collector to obtain data entries from multiple sources, a normalization module to enable the selection and ordering of manipulating modules, and a learning module to apply machine learning algorithms and improve the normalization rules based on the model derived from the data entries. The system can carry out the selection and ordering process either manually by a human user or automatically by a computer processor based on the normalization rules.
Overall, this patent presents a method and system for normalizing automotive data obtained from different sources and in different formats. By applying machine learning algorithms and utilizing manipulating modules, the method ensures that the data entries are transformed into a uniform format, allowing for easier analysis and use in various automotive data consumer software applications.