Bayesian statistics: principles and benefits

http://library.wur.nl/frontis/bayes/03_o_hagan.pdf

This article is meant to summarize the basics of Bayesian statistics for beginners.

In Bayesian statistics:

• a statistical model is created to link data to parameters
• prior information about parameters is formulated
• inferences about parameters are derived from the resulting posterior distribution

Graphically, the narrower the curve, the tighter the parameters. The difference between frequentist methods and Bayesian analysis is the use of past information, which is principally subjective. It is important for the prior information to be defensible and reasonable. The author believes subjectivity is a strength of the system because it allows for the examination of posterior distributions from different informed observers.

Until the 1990s, computational tools for conducting Bayesian analysis were nascent or non-existent. While there are tools for the specialist available today, the general practitioner of Bayesian analysis will find there are few user friendly tools available.

The author enumerates a number of benefits for using Bayesian analysis. They are:

• It provides meaningful and intuitive inferences.
• It can answer complex questions cleanly and exactly.
• It makes use of all available information.
• It is well suited for decision-making.
  

Gap analysis methodology for identifying future ICT related eGovernment research topics – case of “ontology and semantic web” in the context of eGover

http://ecom.fov.uni-mb.si/proceedings.nsf/Proceedings/874F789164F6DD6DC12572EE007A6BB2/$File/Paper95.pdf

This study evaluates the the gaps between the current state of government e-services to those proposed by numerous European conventions. The authors review how they defined and identified those gaps, how they were interrelated, and the methodology used to draw conclusions.

The use of information and communication technologies (ICT) currently employed by governments is poor. The future needs demands placed on governments to meet constituent demands require the streamlined implementation of ICT. The results of this study are part of the eGovRTD2020 project for which this work was commissioned.

For the sake of the study, a gap was identified as either expressing a mismatch between issues of consideration, or an issue of research currently not under consideration. There are five main steps the authors undertook to identify gaps and actions toward future scenarios. Generally speaking, they are:

• where are we at?
• where can we go?
• how do we get there from here?
• this is how we get there.
• this is what we need to do to get there

The researchers then grouped the gap areas together and conducted conclusions on how they were interrelated (the later via a SWOT-esque approach).

The authors highlight four key steps they considered while conducting their gap analysis. They are:

• understand the current environment
• understand the broader context of the environment (for a holistic view)
• base the results on a clear assessment framework
• support the analysis quantitatively
  

Game Theory, Political Economy, and the Evolving Study of War and Peace

By Bruce Bueno de Mesquita

http://www.apsanet.org/imgtest/APSRNov06BuenoDeMesquita.pdf

Summary:

Studies of war and peace increasingly center around domestic interests and institutions for clues on how to shape international affairs. This change in thinking coincides with advances made in non-cooperative game theory and political economy modeling.

De Mesquita refutes realism and state-centric theories as logical explanations for the causes of war. He simultaneously enhances the validity of the liberal peace theory by examining the influences which democrats need to consider when threatening or declaring war.

Realism's claim about a balance of power needed to maintain international stability is refuted by the political economy approach. Simply put, the political economy approach states that the causes of and solutions to international conflict can best be understood by looking within states. It treats leaders as the object of study, not the states as realism does.

A game-theoretic focus concludes that war conducted by rationally acting states is always ex post inefficient. Leaders conduct wars at times to maintain a hold on power, since their domestic constituencies would likely vote them out of office (for democratic states). Autocracies have the advantage of not needing to concern itself with the well-being of their citizenry since they do not face election.

Game theory also validates the liberal peace theory. It claims that leaders (as the object of the political-economy approach to international relations) will only wage war when the outcome is victory, as a 93% success rate for wars initiated by democracies indicate. Since both sides need to consider their reelection prospects, a negotiated settlement to the conflict is the preferred method for resolving conflicts between democratic nations.

Game Theory

http://msl1.mit.edu/ESD10/block4/4.4_-_Game_Theory.pdf

Summary:

Game theory is the branch of decision theory concerned with interdependent decisions. Participants, or players, in a situation, or game, compete to achieve objectives with similar resources. The goal of game theory is not necessarily to "win," but rather to identify an optimal strategy. The sequence of moves a player uses is called a "strategy," and does not need to be wholly unique.

There are two categories of games: sequential and simultaneous. In sequential games, participants take turns acting. The basic strategic rule when applying game theory in this situation is to "look ahead and reason back." This starts with the last decision to be made, then working back through the problem to choose the course of action the other player would make until an initial decision needs made.


Simultaneous games require a more robust analysis due to their overall complexity. There is no "final move" per say in these games. In simultaneous games all possible combinations must be laid out. Players in this type of game should identify their dominate strategy and dominated strategy, the later should never be employed. Thus, the overall range of choices can be limited. Players then can go back to assess if there is another dominate strategy they can employ from the remain outcomes, as well as determine if there is another dominated strategy. This sequence can be replicated until a final strategy is reached, or if a dominate strategy is no longer evident.


In simultaneous games, comptitors without a clear dominate strategy should seek out equalibrium in relation to the other. Equilibrium is reached when both competitors do not have an incentive to change strategy.

Reflections of a Red Team Leader

http://usacac.army.mil/CAC/milreview/English/MarApr07/Craig.pdf
by Susan Craig

Summary:

This article focuses on the military aspects of red team leadership and characteristics commonly found in leaders of red team groups. The author suggests these traits will help any person in an organization think more critically of the environment the decision will be implemented in.

Legislation was enacted to implement red teams to prevent "failures of imagination and critical thinking apparent in the wake of 9/11 and the invasion of Iraq." According to Craig the most important aspect of being a red-teamer is the ability to ask good questions. Those questions should not alienate, but rather stimulate thought and point out assumptions or factors not being addressed. A red team should also identify poor measures of effectiveness and think about better indicators for feedback.

Specifically in terms of the military application of red-teaming the leader needs to consider coalition partner's constraints, capabilities, and political will, as well as their shared opponents.

Craig believes the understanding of cultural idiosyncrasies is forefront to leading a red team. Especially important is the understanding of formal and informal economies; sociological political and religious systems; sociolinguistics; semiotics; and its concept of violence.

More generally applied lessons from leading red teams is espoused upon by Craig. First, the red team should act more as a historian than an analyst (focusing on broad questions as opposed to a very specific question). Red team's insights should be tailored to the audience it addresses. The red team should be as diverse as possible. The implementation of red team recommendations needs leadership committed to making changes to the original plan. The red team leader should possess the abilities to advocate and persuade.

Red Teams: Toward Radical Innovation

http://www-935.ibm.com/services/us/imc/pdf/gt510-6190-red-teams.pdf

Summary:

This executive technology report was produced by IBM and therefore represents a competitive intelligence aspect of red-team analysis.

The executive summary of the report summarizes the main points the best:

Red teams assume the role of outsider to challenge assumptions, look for the unexpected alternatives and find the vulnerabilities of a new idea or approach. By consciously working to assume another perspective and out-do the standard team, they provide one means to getting "out-of-the-box" views and insights.

The report also lists a number of key benefits of a red team:

• Identify significant vulnerabilities
• Discover new uses for innovations
• Challenge taboos and assumptions
• Provide a minority report on a new concept or idea
• Reveal the consequences of different perspectives

Red teams impact all three levels of planning (strategic, tactical, operational). To have any impact the red team must have a proper composition, have the support of management, have an operational relationship with the blue team, have established goals, and have access to most of the information of the blue team.

The report does outline some drawbacks from using a red team approach. The red team needs to take their assignment seriously. At times the red team was marginalized by the red team. Team membership quality lacked to fully exploit the role of adversary. Also, red teams did not receive enough credible information to act appropriately.

An introduction to decision tree modeling

http://www.udel.edu/chemo/SDB/~pdf_papers/JChemo_18_275_2004.pdf

Summary:

The article offers a critique of decision tree mapping as a technique used to generalize data sets.

"In its simplest description, decision tree analysis is a divide-and-conquer approach to classification. Decision trees can be used to discover features and extract patterns in large databases that are important for discrimination and predictive modeling." It most common use is when using exploratory data and predictive modeling applications.

Decision mapping's advantages include recognizing the interpretability of the constructed model, as well as determining inter-dataset relationships. It takes the form of a hierarchical model formed by decision rules represented by nodes. The first node is reffered to as the branch node, with subsequent nodes reffered to as leaf nodes.

The general consesus is to overgrow a decision tree by incorporating all relevent criterion. The tree can then be "pruned" to reduce complexity. Generalizability is enhanced by incorporating ensemble methods such as bagging (randomly selecting samples) or boosting (reweighting criterion).

The Incident Decision Tree: Guidelines for Action Following Patient Safety Incidents

http://www.ahrq.gov/downloads/pub/advances/vol4/Meadows.pdf

Summary:

An Incident Decision Tree (IDT) was formulated to address patient safety issues arising in the United Kingdom. The IDT was developed by the National Health Service (NHS) in response to the alarming number of suspensions incurred by medical staff when patient safety issues arose.

The IDT provides human resource decision makers with a tool for determining the correct line of action when dealing with personnel in the wake of a patient safety issue. The overwhelming course of action before the IDT's implementation was a suspension for the health practitioner. The IDT proposes several different options for managers to consider after working their way through the tree. Applying the IDT allows the manager to implement fair and consistent actions resulting from patient safety mishaps.

The decision maker must think through the system and organization variables in the management of error from the prompts displayed throughout the tree. There are four consecutive trees a manager must consider about the situation. The first is the deliberate harm test (harm is intended), the incapacity test (the practitioner is sick, on medication, or has substance abuse problems), the foresight test (uncertain about protocal, protocal is unstated or poorly adhered to), and the substitution test (an equal would have acted differently). Each subsequent test is applied after the previous test was deemed inappropriate.

The pilot use of the IDT resulted in fewer suspensions for health care practitioners. Also, this lessened to a certain extent the "blame culture" associated with health care providers about self-reporting patient safety issues. Managers focused more on the "what" of the situation, rather than the "who."

A certain weakness was noted when human resource decision makers used the paper version of the IDT. Some decision makers chose suspension as the course of action and then worked backwards through the tree to justify their course of action.

Cultivating Expertise In Informal Reasoning

http://www.philosophy.unimelb.edu.au/reason/papers/CJEP_van_Gelder.pdf

Summary:

Argument mapping technology was introduced in a quasi-study of informal reasoning of first year university students in Australia. Operating under the premise that informal reasoning "appears at quite an early age and continues to develop through secondary and tertiary education...few people manage to become highly proficient." Informal reasoning was likened to a social tennis player: without dedicated practice and guidance, the athlete will never reach full potential. Thus, after initial training "experience in a domain is not related to the level of expertise."

Van Gelder argues informal reasoning skills are achieved to the extent that one engages in large amounts of deliberate practice...to go beyond ordinary competence. He postulated an increase in informal reasoning practice would have a positive effect on overall reasoning skills. To test his postulation, he utilized argument mapping software as a method to used in conventional logic instruction. The main benefit of using argument maps was to "enhance the quantity and quality of feedback" in reasoning activity.

The results of his study indicate students made substantial gains in informal reasoning skills. Participants gained approximately as much over one semester as they would have over 3-4 years of university study. Also, the amount of gain was positively to amount of practice.

The author does admit the results do not establish that deliberate practice is necessary for advanced expertise in informal reasoning.

Dialectic and Method in Aristotle

Aristotle. Image via Wikipedia

http://aristotle.tamu.edu/~rasmith/dialectic-and-method.pdf

Summary:

Aristotle's overall method is described as a philosophical inquiry beginning with appearances and undertaking to resolve apparent puzzles. Dialectic arguments originate from endoxa, or commonly held opinions. The enoxa serve as the first series of premises of scientific demonstrations. Arguements are classified according to their premises: some are in accordance with a particular art (field of study), i.e. rest on premises peculiar to that art, whereas others are general. It is the general arguements which are available for dialectic inquiry; the focus is on deriving the general truth as opposed to circumstancial truth.

Aristotle came to see the inadequacy of of the appeal to intuition for the justification of arguements and sought its replacement through dialectical proofs. To establish truth, a method allowing participants to syllogize from common beliefs needed establishment. This is known as the dialectic.

Since Aristotle believed everyone has a built-in grasp of the truth, the opinions of the many, as well as the wise are acceptable, although each needed clarification and correction. The dialectician is to collect the views from each type of person and use them to gauge the acceptability of premises to a particular opponent. However, not everyone's opinion is treated with equal weight.

When a general agreement is initially reached when attempting dialectic dialogue, it is acceptable in some instances to perpetuate false premises for examination. The false premise than can be argued for and against to establish truth in dialectic dialogue. The key property of dialectic is to examine completely the opinions presented on the topic from the truths contained therein.

DIALECTICS IN KNOWLEDGE ORGANIZATION: AN APPROACH

http://www.inf-wiss.uni-konstanz.de/infwiss/download/isi1998/1_isi98-dv-curras-madrid.pdf

Summary:

"Dialectics" comes from the Greek word "dialogue" which means "to converse between." Thus dialectics is the technique of conversing. Definitions of dialectics changed throughout the years to mean "the study of the theory of knowledge," as well as the art of "demonstrating a proposed thesis by means of the classification of concepts and the rigorous distinction between them.

In reality, dialectics is, in general, a means of organizing knowledge consisting of proposing ideas by putting forward their opposite and reasoning (in dialogue) to obtain a result which is assumed to be the truth. The process involves forming a thesis (what is), its antithesis (what could be), and the synthesis (what it becomes).

In our current age dialectics has been divided into two categories. Scientific dialectics deals with applying intuitive-rational knowledge, object-subject, and individual-society dichotomies against each other. Systems dialectics applies to all branches of science due to the ever increasing changes in our globalized world.

The act of thinking and thought is linked to dialectics insofar as the latter is a means of discerning between true and false (true thought, false thought or the wrong hypothesis) to understand things. The result is the subjective human truth.