MIS 696A Final Presentation

Victor Benjamin, Joey Buckman, Xiaobo Cao, Weifeng Li, Zirun Qi, Lee Spitzley, Yun Wang, Rich Yueh

Agenda

• Introduction•Literature review•Research gaps•Data collection•Results•Conclusion

Introduction

•What is MIS?•Past work vs. current work•Conference papers•Lower submission-to-acceptance time

MIS

Decision Support/Communication Organizational Behavior Electrical

Engineering/Healthcare Economics/Accounting

Decision Support Behavioral Economics Healthcare and Engineering

-Information Retrieval/Artificial Intelligence

- Collaboration-Human-Computer Interaction-Social Issues and Ethics

-Economics/Decision Science-E-commerce

-Databases-Software Development and Engineering-Telecommunications

Timeline of Greatness

• Artificial Intelligence and Information Retrieval• 1967: The successful knowledge-

based program Dendral is built• 1995: Bayesian method developed

for determining atomic positions• 1999: EcoCyc is built to query and

explore the genetics of E.Coli

Timeline of Greatness

• Collaboration• 1971: Delphi method proposed as

group communication structure• 1987: Foundation for the study of

GSS• 1991: Benefits and drawbacks of

GSS established• 1996: Groupware Grid proposed

Timeline of Greatness

• Database• 1961: IDS developed• 1970: E.F. Codd published article on

relational technology• 1973: Michael Stonebraker developed

Ingres/IBM with System R• 1976: Peter Chen and ER Model• 1980: First database build on Oracle

SQL

Timeline of Greatness

•Decision Science•1980: Framework for DSS developed by Ralph Sprague•1987: Foundation of GDSS•2000: Proof of 2-Player Zero Sum Game Equilibrium

Timeline of Greatness

• Economics• 1985: Pricing of computer services• 1988: Switching costs and lock-in

theories• 1993: Productivity paradox of IT• 1996: Emergence of E-Commerce• 1999: Economics of global IT

Timeline of Greatness

• Human-Computer Interaction• 1952: Englebart begins defining information

manipulation problems• 1962: Licklider outlines “Man-Computer

Symbiosis” goals• 1965: First “computer mouse” unveiled (SRI)• 1977: Xerox PARC explores WYSIWYG

displays• 1977: “ZOG: A Man-Machine Communication

Philosophy” (CMU)

Timeline of Greatness

• Social Issues and Informatics• 1988: Adaptive technology required for

team differences• 1994: Fair use and digital data• 1998: Trust in global virtual teams• 2000: Framework to study technology in

organizations• 2001: Intellectual property in an open

information environment

Timeline of Greatness

• Systems Analysis and Design• 1968: Systems Analysis became a formal

discipline• 1972: Information hiding was promoted• 1979: “Structured Design” was published by

Edward Yourdon• 1980: Workflow emerges• 1986: Introduction of object-oriented

development• 1997: UML 1.1 was submitted

MIS Hall of Fame

Hsinchun Chen

University of Arizona

MIS Hall of Fame

Andrew B. Whinston

University of Texas at Austin

MIS Hall of Fame

Ronald E. Rice

University of California Santa Barbara

MIS Hall of Fame

Izak Benbasat

University of British Columbia

MIS Hall of Fame

Jay F. Nunamaker

University of Arizona

Research Gaps

• Previous groups examined journal papers• Journal papers have a lag time of

1 – 2 years•Good indicator of long-term

trends and historical information

Our Focus

•Conference papers•A good indicator of where the field is today•A predictor of the near future•Larger quantity of papers

Data Collection

•Conference papers are not easy to collect (not in a single database)•Some conferences cannot be collected online

Data Collection

•After a search of each database we decided to collect: ICIS, AMCIS, CSCW, HICSS, KDD, SIGIR, WWW •Time Frame: 2008-2012

Data Collection

•Data:•Title, Abstract, Keywords of total 6,036 papers.

Data Collection

•Different conferences have different website patterns

(HTML tags, structures etc.)•We programmed text scrapers in Python and PHP for different website patterns

Data Collection

•Data Sources• Microsoft Academic Search• ACM Proceedings• Official Conference Websites

Data Collection

•Source Comparison

  Microsoft Academic

ACM Proceedings

Official Websites

Data Amount Most conferences missing recent years’ paper

Cover all ACM conferences

One conference for each site

Data Quality Detailed info Missing key words

Detailed info

Data Access API + Json parsing

HTML Crawler + Information Extraction

HTML Crawler + Information Extraction

Data Collection

•Microsoft Academic – JSON Parsing

Data Collection

• ACM Proceedings IE – Regular Expression

Data Collection

•Results

What is LDA?

• Latent Dirichlet Allocation•Algorithm used for topic modeling•Originates from computer science in

2002• Some open source tools exist to help

researchers employ LDA

How LDA works

• Computes chance of certain words appearing together• Also looks for word groups that appear exclusive from

one another• Assumes each document can be a mixture of various

topics• Returns clusters of words to user and topical make-up

of each document

An example of LDA

• Suppose you have the following set of sentences:• I like to eat broccoli and bananas.• I ate a banana and spinach smoothie for breakfast.• Chinchillas and kittens are cute.• My sister adopted a kitten yesterday.• Look at this cute hamster munching on a piece of broccoli.

• According to LDA• Sentences 1 and 2: 100% Topic A• Sentences 3 and 4: 100% Topic B• Sentence 5: 60% Topic A, 40% Topic B• Topic A: 30% broccoli, 15% bananas, 10% breakfast, 10% munching, … (at

which point, you could interpret topic A to be about food)• Topic B: 20% chinchillas, 20% kittens, 20% cute, 15% hamster, … (at which

point, you could interpret topic B to be about cute animals)

Example source: http://blog.echen.me/2011/08/22/introduction-to-latent-dirichlet-allocation/

An example of LDA

• Suppose you have the following set of sentences:• I like to eat broccoli and bananas.• I ate a banana and spinach smoothie for breakfast.• Chinchillas and kittens are cute.• My sister adopted a kitten yesterday.• Look at this cute hamster munching on a piece of broccoli.

• According to LDA• Sentences 1 and 2: 100% Topic A• Sentences 3 and 4: 100% Topic B• Sentence 5: 60% Topic A, 40% Topic B• Topic A: 30% broccoli, 15% bananas, 10% breakfast, 10% munching, … (at

which point, you could interpret topic A to be about food)• Topic B: 20% chinchillas, 20% kittens, 20% cute, 15% hamster, … (at which

point, you could interpret topic B to be about cute animals)

Example source: http://blog.echen.me/2011/08/22/introduction-to-latent-dirichlet-allocation/

Conclusion

• Trending toward technical research•Away from behavioral research•No more TAM• Future work: broaden range of

conferences