Comp Sci Intelligent agent doubt-information systems

  • Thread starter Thread starter shivajikobardan
  • Start date Start date
  • Tags Tags
    Systems
AI Thread Summary
The discussion centers on the design criteria for intelligent agents in executive-level software, emphasizing the need to understand executives' specific information requirements. It critiques the use of focus groups, suggesting they may introduce bias in defining these criteria, and advocates for individual interviews with key users to gather unbiased insights. The methodology described in the referenced study is viewed as potentially favoring predetermined designs over genuine user needs. The importance of establishing clear minimum and specific requirements through structured interviews is highlighted. Overall, the focus remains on ensuring that software design aligns closely with actual user demands rather than preconceived notions of an "ideal" agent.
shivajikobardan
Messages
637
Reaction score
54
Homework Statement
what is minimum requirement vs specific requirements to be called in as intelligent agent for any software code piece?
Relevant Equations
none
https://uobrep.openrepository.com/b...d=16B243A8F6C1FC1242A75969D18D00B7?sequence=5

here is some information.

1646847632266.png


I am unable to figure out which one is minimum requirement??and which one is specific requirement.
 
Physics news on Phys.org
From the abstract:
The study adopted an inductive approach using focus groups to generate a preliminary set of design criteria of “ideal” agents. It then followed a deductive approach using semi-structured interviews to validate and enhance the criteria.

Minimum design criteria for executive level software agents includes thoroughly understanding the executives' requirements, and specifically, what information the execs expect to receive from the agents.

Structured interviews with principal and actual users provides data to establish scope, limits and objectives of the basic software design. Conducting focus groups prior to establishing these basic criteria introduces potential bias in later respondents.

I prefer individual interviews with key players, followed by interviews with small groups of executive users, followed by group meetings. Focus groups and group questionnaires may proceed concurrently with interviews with the responses to initial interviews used to tune; that is, focus discussion. Again, the object remains to avoid biasing requirements development to fit a presupposed 'ideal'.

The methodology outlined in the introduction appears ideal for software engineers to sell a preordained design to the executive committee rather than extract basic requirements from end users, depending on focus group structure and whether the execs develop agent architecture in-house or via contract.
 

Similar threads

Replies
61
Views
9K
Replies
6
Views
6K
Replies
65
Views
10K
Replies
4
Views
8K
Back
Top