According to Ince (2004), There are eleven type of software agents and described as follows:
a. Chatterbots
converses with the user in some restricted form of natural language.
b. Commerce agents
In B2B environment, this agents process invoices, checking arrival of material for process, notifying the product arrival and create financial information.
c. Data management agents
Associated with corpora of data that processing text on web site into a summary.
d. Government agents
Used to search for government regulations or extracting government statistics.
e. News agents
Display the breaking news.
Search for specified news by user.
email the news story to user.
Personalize the newspaper according to users' preference.
f. Newsgroup agents
Sorting out, prioritise and display the posting according to the users keywords.
Search the FAQ list for answers to a specified question. Notify the user of the new information is discussed in newsgroup.
g. Shopping agents
Carry out tasks associated with the accessing of retail sites. It is used to compare prices on the retails sites and auction sites on the internet for the products.
h. Software agents
Carry out tasks associated with software
e.g. Notify for software update
i. Stock agents
Associated with the purchasing and selling stocks and shares.
j. Update agents
Notify the user when a change has occured which is of significance to the user
2. Describe how techniques such as artificial intellgence and statistical techniques are used in software agents.
Fuzzy agents is a kind of artificial intellgence which implements fuzzy logic. This agent uses a rule based system to deal and make decision with the conditions it adopt. (Wikipedia 2009)
In statistical techniques, the software agents uses data mining algorithms to analyse the rules and common behaviours from the information rich environments.
3. Identify various activities in e-commerce where software agents are curently in use.
According to Clurman, W., Foley, T., Guttman, R., Kupres, K. (1997), the following activities are currently in use.
i. Personalization
This is the sense in which it is literally an “agent” taking action in the interests of another just as a real estate agent is understood to represent the interests of a buyer or seller. The outcome of the agent’s primary task is controlled by specific information supplied by the user, the agent is personalized for that user.
ii. Brokering
Brokering is different from personalization in that brokering is the function of information retrieval and delivery, given personalized instructions. Again, the search engine that delivers query results is a simple example of a broker, as it gathers and delivers results based on personalized input. So-called “meta-indexes” perform this function on more than one data source.
iii. Negotiation
Negotiating agents may be empowered to execute transactions on behalf their users. The “program-trading” accepting and reviewing information supplied by the activity of an agent and determine the user threshold to make buying or selling decisions financial markets.
4. Computing ethics and bot programming case study: rocky
a.
An account username: train1 and password: train1 is obtained.
Figure 1. Login Screen
Figure 2. After login
Figure 3. Starting the bot
Figure 4. Normal Option
In the screen of normal option, the bot does not seems to know anything and does not response to the user typing. It always says "I don't understand that."
Figure 5. Say Option
In the screen with say option, the bot response, but it usually give uncertain or unexpect answer or asking unexpected questions.
Figure 6. emote Option
In the screen with emote option, the bot response, the response is much better especially on the postive and negative matter and illness . It still give some uncertain or unexpect answer or asking unexpected questions. When the answer from user is short. Sometimes it gives no response at all.
Figure 7. Stop the Rocky
c.
Accroding to Wikipedia (15 May 2009), ELIZA was a computer program of primitive natural language processing. It uses simple pattern matching techniques to response to its users.
As I chatted with the Rocky about, the Rocky is trying to match the pattern for postiive and negative attitude of the users. It seems not response the long and complex sentance well.
Reference:
Nwana H. S. (1996), Software Agents: An Overview, Retrieved 13 May 2009 from http://www.sce.carleton.ca/netmanage/docs/AgentsOverview/ao.html
Wikipedia (2009), ELIZA, Retrieved 18 May 2009 from http://en.wikipedia.org/wiki/ELIZA
Grossklags J., Schmidt C. (2003), Artificial Software Agents on Thin Double Auction Markets - A Human Trader Experiment, Retrieved 15 May 2009 from http://www.sfb504.uni-mannheim.de/~cschmidt/paper/IAT2003-grossklags-schmidt.pdf
Wroclaw (2007), Using Data Mining Algorithms for Statistical Learning of a Software Agent, Retrieved 17 May 2009 from http://portal.acm.org/citation.cfm?id=1482853
Zhang S., Hu Q., Wang D. (2007), Application of Software Agent to e-Commerce Consumer Buying Support, Retrieved 18 May 2009 from http://ieeexplore.ieee.org/Xplore/login.jsp?url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel5%2F4318334%2F4318335%2F04318860.pdf%3Farnumber%3D4318860&authDecision=-203
Clurman, W., Foley, T., Guttman, R., Kupres, K. (1997), Electronic Commerce with Software Agents, Retrieved 18 May 2009 from http://alumni.media.mit.edu/~guttman/research/commerce/papers/commerce.pdf
xxxxxx OLD ANSWER for Question 3 xxxxxx
i. Supporting
a. Consumer Buying Support
According to Zhang, Hu and Wang (2007), "Facing the flood of information and increasing competitive environment, the e-Commerce seller must make its software facility more customer-oriented, more intelligent, and build one-to-one marketing. A new e-Commerce model is presented to support consumer buying decision making."
b. Health Care Support
Moreno, A., Nealon, J. L. (n.d.) stated that "software agents technology is being applied in very diverse problems in health care, ranging from community care to management of organ transplants."
xxxxxx END xxxxxx
Posts
No comments:
Post a Comment