之前好多课上记的乱七八糟的东西都丢在了博客上，显得很乱，以后笔记可能还是写在ppt上多一些，关于课程的信息和一些总结会丢在博客上，保证博客的质量。

---

CMSC5741 - Big Data Technology and Applications

textbook: Mining of Massive Dataset（已借）

Instructor: Prof. Michael R. Lyu

Tutor: Zeng Jichuan

exam: Nov.6 Midterm exam

Assessment Scheme and Deadlines:

• 20% Assignment
• 40% Midterm examination
• 40% Project : Proposal, Presentation, Report

Backgroud Knowledge: Tensorflow, Amazon EC2

---

CSCI5570 - Large Scale Data Processing Systems

website Account:

• Username : csci5570

• Password : huskydatalab

Instructor: Prof. James CHENG

Tutor: Tatiana Jin

Lecture/Lab:

• Tuesday 13:30 Lecture && Lab
• Wednesday 14:30 Lecture

Assessment Criteria:

• 30% Survey paper : select one topics (DDL: Dec 10, 2018)
• 70% project : deadline: DEC 20

---

CMSC5724Data Mining and Knowledge Discovery

Instructor: Yufei Tao

Tutor: Shangqi Lu

Assessment Criteria:

• 30% Project
• 30% Short Tests (three times in class)
• 40% Final (Open-book)

---

CMSC 5720 - Project I

Instructor: Prof. James CHENG

Options:

1. NN-descent （kn-graph的近似算法）
2. search with fa2ss（facebook的相似性检索库）
3. multiprobe with tree（基于树的哈希方法）
4. LSH for MZPS （lsh）
5. 数据收集->存储->分析 系统
6. topic modeling on ps archetective -> （LDA FlexPS->parameter server拓展）
7. 调度算法，同步/异步 任务，在不同集群下测试算法，分布式，任务的表现
8. 矩阵分解 Distributed MF（矩阵分解） on Actor Framework （nomad,lftf acf或者akka -> cpu to gpu to scheduling）
9. Clustering-aware query (database query optimizer)