INFORMS Nashville – 2016
365
the accident classes (injury or property damage). The year-wise results showed
that RF outperforms other classifiers with higher accuracy, i.e., 92.7%, 91.4%,
and 91.8%, and area under curve values as 0.926, 0.910, and 0.912, respectively
as tested with 10-fold cross validation for years 1, 3, and 4.
2 - Analyzing Analytics Curricula
Thomas Tiahrt, Assistant Professor of Decision Sciences, University
of South Dakota, 414 E. Clark St., Beacom School of Business,
Vermillion, SD, 57069, United States,
thomas.tiahrt@usd.eduThe data deluge has produced a need for analytics professionals across the
organizational spectrum. The academic response has been the establishment of a
host of new analytics programs, combining courses already taught with
innovative offerings, often leveraging existing institutional strengths. We examine
the curricula of one hundred twenty-three analytics programs using text mining
to discover commonality, differences and subdivisions. Our objective is to provide
information for academics either developing or modifying an analytics
curriculum, employers evaluating graduate skills and their needs, as well as
students considering the alignment of coursework with their interests.
3 - When Positive Sentiment Is Not So Positive Textual Analytics And
Bank Failures
Majeed Simaan, RPI, 231 Congress St., TroyY, NY, 12180, United
States,
simaam@rpi.edu, Aparna Gupta, Mohammed Zaki
Looking at 10-K annual reports for a large sample of banks in the 2000-2014
period, 52 public bank holding companies that were associated with bank failures
during the global financial crisis serve as a natural experiment. Utilizing negative
and positive dictionaries proposed by Loughran and McDonald (2011), we find
that both sentiments on average discriminate between failed and non-failed banks
80% of the time. However, we find that positive sentiment contains stronger
predictive power than negative sentiment; out of ten failed banks, on average
positive sentiment can identify seven true events, whereas negative sentiment
identifies five failed banks at most.
WA07
102B-MCC
Data Mining in Marketing
Sponsored: Data Mining
Sponsored Session
Chair: Jiaqi Liu, MD, United States,
buptliujiaqi@qq.com1 - Value In The Eye Of The Beholder: Insights From Data Mining
Diamond Prices.
Stanislav Mamonov, Montclair State University, 1 Normal Ave,
School of Business, Montclair, NJ, 07043, United States,
stanislav.mamonov@montclair.edu, Tamilla Mavlanova,
Nibedita Mohanty, Jacqueline Galal
We use a large dataset of diamonds, currently available for sale, to analyze the
effects of physical properties of diamonds on diamond prices. The results suggest a
high degree of subjectivity in diamond pricing that challenges the accuracy of
predictive models. To illustrate the apparent subjectivity, we discuss several price
anomalies that are present in the dataset.
2 - Paid Search Advertising: Ideas And Pitfalls
Lihui Shi, Centerfield Corporation, 1637 E Maple Avenue, Apt 6,
El Segundo, CA, 90245, United States,
shilihui@uw.eduIn internet marketing, search advertising is a method of placing online
advertisements on web pages that show results from search engine queries. I will
cover search engine marketing (SEM), search engine optimisation (SEO), pay-
per-click (PPC), cost-per-click (CPC), cost-per-impression (CPM), search engine
advertising, sponsored listings, paid for placement, etc. Keyword bidding, search
ads, display ads, product listing ads (PLA), native ads will be discussed in details,
with their definitions, methodologies and challenges. It will provide an overview
of paid search advertising, and how this industry utilizes the statistical methods,
data mining and machine learning techniques.
3 - Research On The Influence Of Microblog Advertising Of Theatre
Chain On Its First Week Box Office Revenue
Jiayin Qi, Shanghai University of International Business and
Economics, Shanghai, 100876, China,
qijiayin@139.com, Jiaqi Liu
With the participation of enterprise in social network, a question arises of how
enterprise can gain more real profit from social network advertising. This article
analyzes the relationships among contents of enterprise’s online post, consumer’s
participation on microblog, and short-term product sales. Taking movie market as
research set, this research finds that short-term product sales is positively
influenced by the information quantity of enterprise’s post, and the influence is
mediated by the “repost” of other users of Sina microblog. The mechanism
examined in this research on the influence of microblog advertising on short-term
sales can be guidelines of social network advertising.
WA08
103A-MCC
Supply Chain Mgt, Green
Contributed Session
Chair: Cao Binbin, PhD Candidate, School of Management, Xi’an
Jiaotong University, No.28, Xianning Road, Xi’an, Shaanxi, Xi’an,
Shaanxi, 710049, China,
caobinbin89@stu.xjtu.edu.cn1 - On The Optimal Control Of Remanufacturing Activities With
Quality Variation Of Returns
Sajjad Farahani, PhD Student, University of Wisconsin -
Milwaukee, Milwaukee, WI, 53211, United States,
farahani@uwm.edu, Wilkistar A Otieno, Xiaohang Yue
This paper develops a model for the optimal disposition decision for product
returns in a remanufacturing system with limited recoverable inventory capacity.
The quality grade of returned products is uncertain and remanufacturing cost
increases as the quality level decreases, and remanufacturer wishes to determine
which returned product to accept to be remanufactured for reselling, and any
unaccepted returns may be salvaged at a value that increases with their quality
level. A continuous time Markov chain defined and Matrix-Geometric method is
applied to evaluate various performance measures for this system and obtain the
optimal remanufacturing policy.
2 - Forwarders’ Cooperative Game In Excessive Capacity Exchange
Xuefei Shi, Southeast University, Sipailou II, nanjing, 211189,
China,
230159160@seu.edu.cn, Haiyan Wang, Minghui Lai
We study logistics forwarders’ cooperative game in capacity exchange. The
forwarders book their capacity in advance, and may have capacity deficit or
excessiveness after their random demand are realized. Thus they can
cooperatively exchange their excessive capacities to improve their profits. We
develop several profit sharing rules.
3 - The Implications Of Environmental Regulations On Market Area
Size And Green Investment Decisions
Nazli Turken, Assistant Professor, Cleveland State University,
11900 Edgewater Dr. Apt 1308, Apt 1308, Lakewood, OH, 44107,
United States,
n.turken@csuohio.eduIn this paper we analyze the effect of different environmental regulations on the
market area size and green technology investment of a company. We try to
identify the environmental regulations that minimize local or global emissions.
We identify scenarios when the green technology and end-of-pipe abatement
decisions are completely separated from each other.
4 - Supplier’s Environmental Innovation Under Government’s
Emission Tax On A Manufacturer
Bosung Kim, PhD Candidate, KAIST College of Business, 85 Hoegi-
ro, Dongdaemun-Gu, Seoul, 130-722, Korea, Republic of,
kim-bs@business.kaist.ac.kr, Se Youn Jung, Kun Soo Park
We study supplier’s environmental innovation in a supply chain under
government’s emission tax. We analyze and compare the equilibrium outcomes
under three types of supply chain contracts: wholesale-price, quality-dependent,
and revenue-sharing contracts. In addition, we discuss the impact of factors that
can discourage the supplier’s innovation.
5 - Joint Decision on Emission And Pricing In Waste-based Supply
Chain: Insights from Hybrid Environmental Policies
Cao Binbin, PhD Candidate, School of Management, Xi’an
Jiaotong University, No.28, Xianning Road, Xi’an, Shaanxi,
710049, China,
caobinbin89@stu.xjtu.edu.cn,Xiao Zhongdong
Carbon Disclosure Project reported around 50% of a product’s emission is
generated from manufacturing process. A more general waste-based supply chain
model including one retailer, one manufacturer and one supplier was built. We
analytically showed joint decision processes on emission level of the
manufacturer and pricing in supply chain under three hybrid policies including
tax and offset, cap and trade, tax and trade. An analytical comparison between
different hybrid policies was provided in terms of profits and emission reduction.
Numerical study was presented to further illustrate the effects of emission
reduction investment factors and parameters of each hybrid environmental policy.
WC08