Course Outline

  1. Distributed data under big data
    1. Data mining method (training single machine + distributed prediction: traditional machine learning algorithm + Mapreduce distributed prediction,)
    2. Apache Spark MLlib
  2. Recommendations and targeted advertising:
    1. The natural language part
    2. Text clustering, text classification (labeling), synonyms
    3. User profile restoration, tag system
    4. Recommendation Algorithm Strategy
    5. Lift between classes, lift within classes, how to be precise
    6. How to build a closed loop for recommendation algorithms
  3. Logistic regression, RankingSVM,
  4. Feature recognition: (deep learning and automatic feature recognition of graphics)
  5. natural language
    1. Chinese word segmentation
    2. Topic Model (Text Clustering)
    3. Text Categorization
    4. Extract keywords
    5. Semantic parser, word2vec to word vector
    6. RNN Long short-term memory (TSTM) Architecture
 21 Hours

Testimonials (1)

Related Categories