Building Topic/Trend Detection System based on Slow Intelligence

1
Building Topic/Trend Detection System based on
Slow Intelligence

• Chia-Chun Shih Ting-Chun Peng
• Institute for Information Industry
• Taipei, Taiwan

Presented at DMS10 special session on Slow
Intelligence Systems
2
Agenda

• Introduction
• Topic/Trend Detection System
• Topic/Trend Detection System with Slow
  Intelligence
• Conclusion

3
Introduction
4
Introduction
Facebook Users
Twitter Posts
Blog Posts

• Social media is prevailing
• Social media is a reflection of real-world
• An experiment from HP Social Computing Lab shows
• Twitter-rate time series can accurately predict
  box-office movie sales with Adjusted R2 0.973
  (amazing!!)
• The emerging market for Social Media Monitoring
  Service
• E.g., Nielsen Buzzmetrics, Radian6

5
Introduction
(contd)

• Topic Detection and Tracking (TDT)
• Initiated by DARPA at 1996
• discover the topical structure in unsegmented
  streams of news reporting as it appears across
  multiple media
• Tasks
• Topic Detection
• Topic Tracking
• First Story Detection
• Story Segmentation
• Link Detection

6
Introduction
(contd)

• Slow Intelligence provides a software development
  framework for systems with insufficient computing
  resources to gradually adapt to environments to
  handle complexities

Environment
Knowledge-based Controller
Problem
Solution
1
2
3
4
Enumerator
Adaptor
Eliminator
Concentrator
Slow Intelligence System
7
Introduction
(contd)

• In this paper, we propose a design of online
  topic/trend detection system for Social Media
  with the advantages of Slow Intelligence.
• Four complexities of designing online topic/trend
  detection systems are identified, along with
  corresponding Slow Intelligence solutions.

8
Topic/Trend Detection System
9
Topic/Trend Detection System

• Objective
• Detect current hot topics and to predict future
  hot topics based on data collected from Social
  Media
• Three components
• Crawler Extractor Collect data and extract
  information from Social Media
• Topic Extractor Detect hot topics from a set of
  text documents
• Trend Detector Detect trends (future hot topics)
  based on currently available data

Current Hot topics
Crawler Extractor
Topic Extractor
Trend Detector
Social Media
Future Hot topics
10
Topic/Trend Detection System
(contd)

• Crawler Extractor

Social Media
HTML documents
Users Keywords of Interests
Web Crawler
Text documents
Web data DB
Topic Extractor
Information Extractor
Extract articles and metadata (title, author,
content, etc) from semi-structured web content
Crawler Extractor
11
Topic/Trend Detection System
(contd)

• Topic Extractor

Web data DB
Current topics
Topic Word Extraction
Topic Word Clustering

• Apply TF-IDF scheme to generate Top-N topic
  words for each document

• Apply clustering algorithm to cluster topic
  words into topic groups. The topic groups are
  treated as topics

Current Hot topics
Hot topic extraction

• Apply aging theory to find hot topics

Topic Extractor
12
Topic/Trend Detection System
(contd)

• Trend Detector

Trend Detector

• The Trend Estimation Algorithm is a black box
  now, however, it will find its way when Slow
  Intelligence is involved in the system

13
Topic/Trend Detection Systemwith Slow
Intelligence
14
T/TD System with Slow Intelligence

• Four complexities of designing online topic/trend
  detection systems
• 1. It is unlikely to collect all web data based
  on limited amount of computing resources. The
  system needs to develop data collection
  strategies which can concentrate limited
  resources on collecting important web data.

Crawler Extractor
15
T/TD System with Slow Intelligence
(contd)

• 2. Many computation methods are available for
  estimating trends. If parameter settings are also
  taken into account, there are too many
  combinations to choose. Furthermore, Internet is
  a changing environment, which means current best
  solution may not perform well in the future. The
  system needs to automatically (or at least
  quasi-automatically) find best solution from many
  alternatives in a changing environment.

Trend Detector
16
T/TD System with Slow Intelligence
(contd)

• 3. The crawler needs to revisit websites to
  collect up-to-date data in hourly or daily
  intervals. Each site has different amount of
  to-be-update data and different policy to
  restrict frequent access, which are unknown
  beforehand. The system needs to find feasible
  data collection schedule based on past experience.

Crawler Extractor
17
T/TD System with Slow Intelligence
(contd)

• 4. Any changes in web pages may disrupt
  Extractors. It needs automatic repair mechanism
  for Extractors if many websites are being
  monitored. The repair mechanism needs to detect
  errors of Extractors, find alternatives, and
  choose the best solution from alternatives to fix
  the disrupted Extractors.

Crawler Extractor
18
T/TD System with Slow Intelligence
(contd)

• 1. SIS to help restrict the range of data
  collection

Knowledge of data
Knowledge of algorithm
19
T/TD System with Slow Intelligence
(contd)

• 2. SIS to help select and adapt trend detection
  algorithms

20
T/TD System with Slow Intelligence
(contd)

• 3. SIS to help scheduling Crawler

21
T/TD System with Slow Intelligence
(contd)

• 4. SIS to help adapt Extractors

22
Conclusion
23
Conclusion

• An online trend detection system requires careful
  resource allocation and automatic algorithm
  adaptation to process huge size of heterogeneous
  data.
• This research adopts Slow Intelligence, which
  provides a framework for systems with
  insufficient computing resources to gradually
  adapt to environments, to response the
  challenges.
• Four Slow Intelligence subsystems are proposed,
  and each subsystem targets a challenge in
  designing online topic/trend detection systems.

24
If you have any questions, please e-mail us

• chiachun_at_iii.org.tw (Chia-Chun Shih)
• markpeng_at_iii.org.tw (Ting-Chun Peng)