“Calculus of concepts”的意思、由来-开放百科全书

The calculus of concepts is an abstract language and theory, which was developed to simplify the reasons behind effective messaging when delivered to a specific target or set of targets. The theory aims to maximize the likelihood of desired outcomes, by using messaging elements and techniques while analyzing the delivery mechanisms in certain scenarios.^[1] The reduction of uncertainty, but not its elimination, is often cost effective and practical.

Empowered by the internet of things (IoT)^[2] the framework looks at numerous device such as smart phones, tablets, laptops, hand held gaming devices, GPS devices, automobile Event Data Recorders and other electronic devices as remote sensors capable of providing data channels.^[3]

By using elements, the theory discovers underlying key concepts and their relations to better understand how messages can by used to elicit desired behaviors through mental model heuristics and biases.^[4] The framework does not serve up spam to potential consumers; it is a new paradigm for effective messaging. The nature of the data produced and consumed by devices in the IoT naturally lends itself to location-based awareness.^[5]

Just-in-time and real-time broadcasting of key messages gives the framework an extra dimension, putting it at the forefront of behavioral methodologies. Broadcasting can take place across a number of platforms, text, photographic, video, audio or even direct human contact.

The use of anchoring-and-adjustment, framing and representativeness heuristics provides fertile grounds for “re-wiring” the decision making processes to include either positive or mitigating mental models of a given concept or set of related concepts.^[6] The “re-wiring” will often produce results that have a significant impact on later decisions and behaviors on the target audience. The framework analyses key factors that influence the effectiveness of messaging mechanisms and how differing approaches can lead to entirely different results.^[7]

Background

The calculus of concepts framework has been practically implemented utilizing a combination of Naive Bayes classification^[8] and Support Vector Machines (SVM) algorithms to actively identify the key components of a messaging campaign and its effectiveness. The effectiveness of a communications campaign is often measured by numerous results including reach, frequency and duration.^[9]

The training data set for the model implementation utilized the potential messages and delivery mechanisms with Actors, Actions, Objects, Contexts and Indicia as a few examples.

Each concept within the framework is treated by the practical implementation as either an independent or dependent variable (as applicable) and therefore may have a meaningful effect on the outcome of any communication. As with any machine-learning tool the Calculus of Concepts model implementation inputs can be either nominal or ordinal and depending on the particular case.

Practical example

Between 2005 and 2012 one of the largest oil companies in China attempted to buy the twelfth biggest oil company in Canada. Initial proposals and takeover plans were rejected due to a number of issues surrounding political tensions.

Stakeholders identified Environment, Context, Domain, Event, Condition, State, Decision, Relation, Actor, Action and Object concepts that needed to be in place to have the key decision makers utilizing the mental models needed to secure the takeover.

Over the next 7 years messaging activities were fielded by the Chinese company and its authorized agents, specifically designed to elicit Ideations and Decisions that would result in the takeover going through. In 2012, the takeover was completed after a coordinated and concerted field messaging activity.^[3]

References

1. ^{{cite book|last=Cialdini|first=Robert B.|title=Influence : the psychology of persuasion|year=2007|publisher=Collins|location=New York|isbn=978-0061241895|edition=Rev.}}
2. ^{{cite web|last=Ashton|first=Kevin|title=That 'Internet of Things' Thing - RFID Journal|url=http://www.rfidjournal.com/articles/view?4986|publisher=RFID Journal|accessdate=30 April 2013}}
3. ^¹{{cite web|title=Calculus of Concepts Case Study - February 2013|url=http://www.hubrisanalytics.com/case-studies/CaseStudyFeb2013.pdf|publisher=Hubris Analytics|accessdate=30 April 2013|archive-url=https://web.archive.org/web/20130228034548/http://www.hubrisanalytics.com/case-studies/CaseStudyFeb2013.pdf#|archive-date=2013-02-28|dead-url=yes|df=}}
4. ^{{cite book|last=Klien|first=Gary|title=Streetlights and Shadows: Searching for the Keys to Adaptive Decision Making|year=2011|publisher=MIT Press|isbn=978-0262516723}}
5. ^{{cite web|last=Associati|first=Casaleggio|title=The Evolution of Internet of Things|url=http://www.casaleggio.it/pubblicazioni/Focus_internet_of_things_v1.81%20-%20eng.pdf|publisher=Casaleggio.it|accessdate=30 April 2013}}
6. ^{{cite book|last=Lopes|first=Lola L.|title=The Rhetoric of Irrationality}}
7. ^{{cite book|last=Holiday|first=Ryan|title=Trust Me, I'm Lying: Confessions of a Media Manipulator|year=2012|publisher=Portfolio Hardcover|isbn=978-1591845539}}
8. ^{{cite web|last=Zhang|first=Harry|title=The Optimality of Naive Bayes|url=http://www.cs.unb.ca/profs/hzhang/publications/FLAIRS04ZhangH.pdf|publisher=University of New Brunswick|accessdate=30 April 2013}}
9. ^{{cite journal|author1=Corinna Cortes|author2=Vladimir Vapnik|title=Machine Learning|date=September 1995|volume=20|issue=3|url=https://link.springer.com/article/10.1007%2FBF0|accessdate=30 April 2013|doi=10.1007/BF0|doi-broken-date=2019-03-14}}

Background

Practical example

References

Further reading