Presentation is loading. Please wait.

Presentation is loading. Please wait.

Crowdsourcing Insights with Opinion Space Ken Goldberg, IEOR, School of Information, EECS, UC Berkeley.

Similar presentations


Presentation on theme: "Crowdsourcing Insights with Opinion Space Ken Goldberg, IEOR, School of Information, EECS, UC Berkeley."— Presentation transcript:

1 Crowdsourcing Insights with Opinion Space Ken Goldberg, IEOR, School of Information, EECS, UC Berkeley

2

3

4

5 “We’re moving from an Information Age to an Opinion Age.” - Warren Sack, UCSC

6 Motivation Goals Engage community Understand community – Solicit input – Understand the distribution of viewpoints – Discover insightful comments Goals of Community Members Understand relationship to other community members Participate, express ideas, and be heard Encounter a diversity of viewpoints

7 Motivation Classic approach: surveys, polls Drawbacks: limited samples, slow, doesn’t increase engagement Modern approach: online forums, comment lists Drawbacks: data deluge, cyberpolarization, hard to discover insights

8 Approach: Visualization

9 Approach: Level the Playing Field

10 Approach: Wisdom of Crowds

11 Related Work: Visualization Clockwise, starting from top left: Morningside Analytics, MusicBox, Starry Night

12 Related Work: Politics Clockwise, starting from top left: EU Profiler, Poligraph, The How Progressive Are You? quiz

13 Related Work: Opinion Sharing Polling & Opinion Mining – Fishkin, 1991: deliberative polling – Dahlgren, 2005: Internet & the public sphere – Berinsky, 1999: understanding public opinion – Pang & Lee, 2008: sentiment analysis Increasing Participation – Bishop, 2007: theoretical framework – Brandtzaeg & Heim: user study – Ludford et al, 2004: uniqueness & group dissimilarity

14 Related Work: Info Filtering K. Goldberg et al, 2001: Eigentaste E. Bitton, 2009: spatial model Polikar, 2006: ensemble learning

15 Opinion Space: Live Demonstration

16

17 Six 50-minute Learning Object Modules, preparation materials, slides for in-class lectures, discussion ideas, hand-on activities, and homework assignments.

18 To try it: google: “opinion space” contact us: http://goldberg.berkeley.edu

19 Dimensionality Reduction low variance projectionmaximal variance projection

20 Dimensionality Reduction Principal Component Analysis (PCA) Assumes independence and linearity Minimizes squared error Scalable: compute position of new user in constant time

21 Canonical Correlation Analysis 2-view PCA Assume: – Each data point has a latent low-dim canonical representation z – Observe two different representations of each data point (e.g. numerical ratings and text) Learn MLEs for low-rank projections A and B Equivalently, pick projection that maximizes correlation between views z z x x y y Graphical model for CCA x = Az + ε y = Bz + ε z = A -1 x = B -1 y

22 CCA on Opinion Space Each user is a data point – x i = user i’s responses to propositions – y i = vector representation of textual comment Run CCA to find A and B, use A -1 to find 2D representation Position of users reflects rating vector and textual response Ignores ratings that are not correlated with text, and vice versa Given text, can predict ratings (using B) z z x x y y Graphical model for CCA x = Az + ε y = Bz + ε z = A -1 x = B -1 y

23 Multidimensional Scaling Goal: rearrange objects in low dim space so as to reproduce distances in higher dim Strategy: Rearrange & compare solns, maximizing goodness of fit: Can use any kind of similarity function Pros – Data need not be normal, relationships need not be linear – Tends to yield fewer factors than FA Con: slow, not scalable δ ij i j d ij i j

24 Kernel-based Nonlinear PCA Intuition: in general, can’t linearly separate n points in d < n dim, but can almost always do so in d ≥ n dim Method: compute covariance matrix after transforming data into higher dim space Kernel trick used to improve complexity If Φ is the identity, Kernel PCA = PCA

25 Kernel-based Nonlinear PCA Pro: Good for finding clusters with arbitrary shape Cons: Need to choose appropriate kernel (no unique solution); does not preserve distance relationships Input dataKPCA output with Gaussian kernel

26 Stochastic Neighbor Embedding Converts Euclidean dists to conditional probabilities p j|i = Pr(x i would pick x j as its neighbor | neighbors picked according to their density under a Gaussian centered at x i ) Compute similar prob q j|i in lower dim space Goal: minimize mismatch between p j|i and q j|i : Cons: tends to crowd points in center of map; difficult to optimize

27 Metavid

28

29 Six 50-minute Learning Object Modules, preparation materials, slides for in-class lectures, discussion ideas, hand-on activities, and homework assignments.

30

31 Opinion Space: Crowdsourcing Insights Scalability: n Participants, n Viewpoints n 2 Peer to Peer Reviews Viewpoints are k-Dimensional Dim. Reduction: 2D Map of Affinity/Similarity Insight vs. Agreement: Nonlinear Scoring Ken Goldberg, UC Berkeley Alec Ross, U.S. State Dept

32 Opinion Space Wisdom of Crowds: Insights are Rare Scalable, Self-Organizing, Spatial Interface Visualize Diversity of Viewpoints Incorporate Position into Scoring Metrics Ken Goldberg UC Berkeley


Download ppt "Crowdsourcing Insights with Opinion Space Ken Goldberg, IEOR, School of Information, EECS, UC Berkeley."

Similar presentations


Ads by Google