Page 174 - Introduction to Electronic Commerce and Social Commerce

5.3 Knowledge Management, Intelligent Systems, and Robots 155

Figure 5.6 AskMe’s expert ?
location system Question
ELS

Step 1: Knowledge Repository
Employee submits
question to ELS Solutions, Best Practices
(e-mail,portal)
Step 2

Is
Answer Yes there a
match of problem
Yes and solution?
Step 4
Is No, question
solution published/e-mailed
approvable? No Go to to known experts via
Boss or outside AskMe Enterprise
committee? expert(s) Knowledge Network

Yes
Step 3
No
Can
company’s expert(s)
create a new
solution? In-company
subject-matter experts

that know only their own language to converse, in real time, In January 2016, Mark Zuckerberg, the CEO of Facebook,
with people speaking other languages. announced that his goal is to build in 2016 an AI-based assis-
More complex applications are those that can help in tant to help in his personal and business activities and deci-
answering customers’ questions made in natural languages. sions. Zuckerberg is teaching a machine to understand his
Another area is that of knowledge-based systems (also known voice and follow his basic commands as well as to recognize
as expert systems). These systems can assist people in making faces of people. These are actually capabilities of smart
decisions, and even make decisions on their own. For exam- robots.
ple, such systems can approve or reject companies’ requests
to purchase online (if they are not preapproved, or do not Example: Pitney Bowes Is Getting Smarter with AI
have a credit line). Other examples include the automation of Pitney Bowes Inc. is a U.S.-based global e-commerce solu-
generating online purchasing orders, and expediting fulfill- tion provider in areas such as shipping and mailing products,
ments. Both Google and Facebook are experimenting with location intelligence, customer engagement, and customer
projects that attempt to teach machines how to learn (machine information management. The company powers billions of
learning) and make decisions. Other companies doing the physical and digital transactions annually, across the con-
same (e.g., Toyota); see Markoff (2015). For applications in nected and borderless world of commerce.
the enterprise, see Dodge (2016). Today, prices are depending on the dimensions, weight,
Knowledge-based systems are also important for innova- and packaging of each product. The prices’ calculations cre-
tion (see spigit.com) and they are related to the areas of ana- ate data that are fed into algorithms. The more data are col-
lytics and big data processing. One of the most advanced lected, the more accurate are the calculations. The company
projects in this area is IBM’s Watson Analytics (e.g., see Taft estimates a 25% improvement in accuracy yield from their
2016 and Niccolai 2015). algorithms.