PhD Intern - Advanced ML schema mapping model
Please note you must be a 2nd or 3rd year PhD student to apply.
Adiona develops advanced logistics optimisation algorithms and software. Funded by the Australian Research Council and partnered with UNSW, our technology helps mobile fleets achieve efficient, minimal cost operation and is a practical solution for improving the sustainability of our cities.
We ingest large datasets from various sources such as Enterprise Resource Planning (ERP) systems, which includes information such as addresses, IDs, timestamps, notes, and much more. A significant challenge is to automate the translation of these variable and customised data sets into a format that can be used to analyse the journey of real world products and people.
The goal of this project is to create a ML model that can use text embedding or other methods to identify and categorise many types of unknown data fields into known fields.
This challenge is well known in industries such as data warehousing which uses a variety of schema mapping tools. The researcher will be expected to examine the state of the art in via literature review.
- Working with large data sets.
- Schema detection, matching, or database normalisation knowledge.
- NLP, language modelling, feature learning or other relevant knowledge.
- Relevant tools such as Google BigQuery, Tensorflow, R, Matlab, Word2vec, GloVe, etc
- Team spirit and can-do attitude are a must!
- Ability to work closely with a team and also independently.
Adiona is based in Surry Hills in Sydney just a 4-minute walk from Central station. However, due to recent circumstances this project can be done entirely remotely or a mix.
The intern will receive $3,000 per month of the internship for 3 to 5 months (to be discussed with each candidate), usually in the form of stipend payments.