Intelligent Medical Services Using Big Data

Intelligent Medical Services



Sponsored by

    Student Research Training Program of China


    Xiaotong Zhang, Yuanzhuo Wang

Project Objectives:

    The project aims to build an intelligent medical system using Chinese Wikis (Wikipedia Chinese, Baidu Baike and Hudong Baike) and information in web pages. The application can help patients get medical advice, including medicine information search and recommendation, and help doctors seeing patients and making the prescription by providing patients' medical records and contacting with the patients.

My Work 1

  • Designed and implemented the coarse-grained entity typing model for entites in Wikipedia, Baidu Baike and Hudong Baike, to extract medicine and disease entities.
  • Extracted features of entities in Wikipedia, Baidu Baike and Hudong Baike, including constructing semantic features from free text and generating attribute features from semi-structured text.
  • Implemented a classification module by Support Vector Machine (SVM).
  • Achieved improvement from 85% to 95% of the F1 measure.

My Work 2

  • Participated in designing the fine-grained entity typing module according to the effect of medicines.
  • Implemented the entity linking module by using context similarity.
  • Selected features to define context similarities, including word frequency, abbreviation, alias, and so on.
  • Implemented the random walk module to generate types. The random walk starts from every node in the graph, and walk to another node according to a transfer possibility.
  • Co-authored a patent in China (CN201510033050.4.) (2nd student author).

My Work 3

  • Participated in building attribute extractor of medicine entities from web pages by Conditional Random Field (CRF) and rule-based extractor.
  • Extracted features of training data to train the supervised attribute extractor based on Conditional Random Field (CRF).
  • Assisted in making rules to extract attributes by cascaded finite state automaton.
  • Co-authored a patent in China (No.CN201510071993.6.) (2nd student author).

My Work 4

  • Built a light search engine for the extracted medicines, including building inverted index, query expansion and rank model using vector space model.
  • Implemented the Andriod client to help patients search medicines and information about disease. It can also help doctors to see patients and make the prescription .