\( \def\RR{\bf R} \def\real{\mathbb{R}} \def\bold#1{\bf #1} \def\d{\mbox{Cord}} \def\hd{\widehat \mbox{Cord}} \DeclareMathOperator{\cov}{cov} \DeclareMathOperator{\var}{var} \DeclareMathOperator{\cor}{cor} \newcommand{\ac}[1]{\left\{#1\right\}} \DeclareMathOperator{\Ex}{\mathbb{E}} \DeclareMathOperator{\diag}{diag} \)

From Big Data to Knowledge:

Brain, Twitter, and Beyond

Xi (Rossi) Luo

Brown University
Department of Biostatistics
Center for Statistical Sciences
Computation in Brain and Mind
Brown Institute for Brain Science

Brown University, ADOCH
April 20, 2016

Slides at bit.ly/xl160420

Big Data and Brain

  • 100 million ( $10^{11}$) neurons
  • 100 trillion ( $10^{14}$) neuronal connections

fMRI data: blood-oxygen-level dependent (BOLD) signals from each cube/voxel (~millimeters), $10^5$ ~ $10^6$ voxels in total.

Complex Data: fMRI Studies

Sub 1, Sess 1

Time 1



Sub i, Sess j

Sub ~100, Sess ~4

Large, multilevel (subject, sess, voxel) data
$100 \times 4 \times 300 \times 10^6 \approx 100 $ billion data points

Predicting Cocaine Relapses

Image: Luo et al, 13

Complex Models: Brain Networks

Find brain networks with 1 trillion ( $10^{12}$) possible connections

Image: Liu and Luo, 15

Course: Stat Methods for Big Data

Applied Mathematics + Computer Science + Statistics

Thanks to the support of Office of the Provost, the Sheridan Center, and School of Professional Studeis

Case Study: Twitter

Hillary Clinton

Bernie Sanders

Imge: wikipedia

Excerpt of Data

Jaime Zhang, student in SMBD

Statistical test: $p<10^{-4}$

Jaime Zhang, student in SMBD


  • Handle big data
  • Complex mathematical modeling
  • Extract knowledge from data
  • Course SMBD covers the above points

Thank you!