Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... ·...

19
1 Editors Prof. Amos DAVID & Prof. Charles UWADIA Arts Technology Faith Science

Transcript of Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... ·...

Page 1: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

1

Editors – Prof. Amos DAVID & Prof. Charles UWADIA

Arts

Technology

Faith Science

Page 2: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Transition from Observation to Knowledge to

Intelligence

25-26 August 2016

University of Lagos, Nigeria

Editors

Prof. Amos DAVID

Prof. Charles UWADIA

Page 3: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

87

Actionable Knowledge Discovery: The Analytics

Hierarchy Process Approach

Ikuvwerha L.O., Odumuyiwa, V.T, Ogunbiyi T.D, Uwadia, C.O.,

Abass O.

Ikuvwerha L.O., Odumuyiwa, V.T, Ogunbiyi T.D, Uwadia,

C.O., Abass O.

Department of Computer Sciences,

University of Lagos, Nigeria Abstract: The traditional data mining algorithms and tools stop at mining data that

provide patterns that satisfy the expected technical interestingness. Meanwhile business

people finds it difficult to apply the mined patterns in taking necessary actions to

support their business needs. Therefore, there is the need to move from ordinarily

mining of patterns to discovering actionable patterns (knowledge). One of the major

issue in actionable knowledge is interestingness measure. Interestingness measure can

be generally divided into two categories: objective measure which is based on statistical

strength of the data mining method and the subjective measure which is based on the

user’s beliefs or expectations of the particular problem domain. This research focus is

on subjective measure of interestingness. Analytic Hierarchy Process (AHP) approach

is used to measure the subjective interestingness of the patterns. The three subjective

measure criteria that were used to formulate the AHP model are pattern actionability,

expectedness and novelty. This research uses an illustrative example to demonstrate the

concept of this model. The criteria/threshold in the algorithm of data mining stage is

used as the objective measure while the AHP model is used as the subjective measure

to evaluate and rank the pattern according to their actionability that is its usability. This

research shows that this model is more effective because it considers both the technical

interestingness and the business interestingness. This research also shows that AHP can

be effectively used as subjective interestingness which is the business interestingness.

Keywords: Actionable patters, AHP, interestingness

Page 4: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Actionable Knowledge Discovery: The Analytics Hierarchy Process Approach

88

1. Introduction

One of the central problems of data mining is the discovering of

interestingness and actionable patterns. “Actionable patterns” is

referred to knowledge that end-user (which could be decision –maker)

can act upon or take action on. Most data mining algorithms and tools

stop at the mining and delivery of patterns satisfying expected technical

interestingness. There are often many patterns mined but business

people either are not interested in them or do not know what follow-up

actions to take to support their business decisions. Therefore, it is

important to filter these patterns through the use of some measures

(interestingness) to produce patterns that are actionable and usable to

the end-users. It is very important to understand the overall approach

before one attempts to extract useful knowledge from data and define

good measures of interestingness that would allow the entire process or

system to discover only the useful patterns. Interestingness measure can

be generally divided into two categories: objective measure which is

based on the strength of the statistical method of the data mining criteria

and subject measure based on the user’s beliefs or expectations of the

particular problem domain. The Analytic Hierarchy Process (AHP)

developed by Thomas Saaty for decision making uses a well-defined

mathematical structure of consistent metrics called Pair-wise

comparison matrix to process the inescapably subjective and personal

preferences of an individual or a group in making a decision. With the

use of AHP, it is easier to construct hierarchies or feedback networks,

then a Pair-wise comparison matrix of elements with respect to a

controlling factor is used to derive ratio scales that are then synthesized

throughout the structure to select the best alternative (saty,2008).

Actionable knowledge discovery is selected as one of the greatest

challenges of next-generation knowledge discovery in database (KDD)

studies (Ankerst, 2002; Fayyad, Shapiro, 2003). Mined patterns are

often non-actionable in the existing data mining and knowledge

discovery in database to real user needs. According to Cao & Zhang

(2007) “the traditional data-centered mining methodology could be

complimented by the involvement of domain-related social intelligence

in data mining which leads to domain-driven data mining “Simply

Page 5: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Ikuvwerha L.O., Odumuyiwa, V.T, Ogunbiyi T.D, Uwadia, C.O., Abass O

89

knowing many algorithms used for data analysis is not sufficient for a

successful data mining (DM) and Knowledge Discovery (KD) project.

The major research questions are: how can data and knowledge

satisfying business users’ request be presented; what is generic business

interestingness framework; and how can this framework produce

reliable and trustworthy knowledge. This paper is organized as follow:

section 2 is the related works, section 3 discusses the AHP model,

section 4 look at an illustrative example and section 5 is the conclusion.

2. Related works

Data mining and knowledge discovery is a process that must involve

the user at every stage from the initial stage to the final stage of analysis.

According to Mcgarry (2005) the taxonomy of interestingness

measures is shown in figure 1

Figure1: Measure of interestingness

Dharanik and Gudikandula (2012) used multi-step mining to

discover actionable knowledge. This proposed framework combines

multiple data source, multiple methods and multiple features to obtain

actionable patterns set using existing algorithms. The model could be

used in the banking sector, insurance etc. The limitations of this model

or framework are; the matrix used to determine the actionability of the

patterns in the specified domain is not well defined and the domain

environment and constraints are not considered.

Cao (2012), summarized the extreme imbalances that exist in the

current data mining, which are:

Algorithm imbalance

Pattern Imbalance

Decision Imbalance

Measure of interestingness

Objective

Coverage Support accuracy

Subjective

Unexpected Actionable Accuracy

Page 6: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Actionable Knowledge Discovery: The Analytics Hierarchy Process Approach

90

The paper treats AKD as closed optimization problem.

AKD:=OPTIMAZATION (PROBLEM, DATA, ENVIRONMENT,

MODEL, DECISION)

AKD is a problem-solving Process that transforms business problem

ѱ with problem status t to a problem- solution ф.

Ѱ(./t) ф( ). …….. (1)

Kavitha and Ramaraj (2013), presented a framework that uses

combined mining and composite approach to generate actionable

patterns in terms of rules. The concept from meta- learning that uses

decision theory was used to formulate a utility interestingness measures

(objective and subjective). Zoo and Mushroom data from the University

of California Irvine was used for the experiment.

Amruta and Balachandran (2013), reviewed the four most used

AKD frameworks for business need. These frameworks are:

Postanalysis-interestingness-based AKD

Unified-interestingness-based AKD

Combined-mining-based AKD

Multisource combined- mining- based AKD

Their performance (the numbers of actionable pattern sets) was

evaluated under decision making system using a real time tennis data

set. The multisource combine-mining-based AKD performs better than

the others.

2.1. What is Actionable Knowledge?

The term “actionable pattern” refers to knowledge that can be

uncovered in large complex databases and can act as the impetus for

some action. It is important to distinguish these actionable patterns from

the lower value patterns that can be found in great quantities and with

relative ease through so called data dredging. This highlights the need

to make optimal use of the human in the KDD process for directing the

exploration and evaluating its results — one cannot simply apply some

Page 7: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Ikuvwerha L.O., Odumuyiwa, V.T, Ogunbiyi T.D, Uwadia, C.O., Abass O

91

predefined procedure to distill large volumes of data into high value

knowledge. Rather, KDD tools must provide users with the insight

required to focus the tool’s search processes and with the means to

efficiently evaluate discovered patterns.

Discovering actionable knowledge has been viewed as the essence

of KDD. However, even up to now, it is still one of the great challenges

to existing and future KDD as pointed out by the panel of SIGKDD

2002 and 2003 (Cao & Zhang 2007a, 2007b). This situation partly

results from the limitation of traditional data mining methodologies,

which view KDD as a data-driven trial-and-error process targeting

automated hidden knowledge discovery (Cao & Zhang 2007a, 2007b).

The traditional data mining methodologies do not take into much

consideration the constrained and dynamic environment of KDD; these

methodologies naturally excludes humans and problem domain in the

loop. As a result, very often data mining research mainly aims to

develop, demonstrate, and push the use of specific algorithms while it

runs off the rails in producing actionable knowledge of main interest to

specific user needs.

2.3.1 Measuring Knowledge Actionability

Often mined patterns are non-actionable to real needs due to the

interestingness gaps between academia focused-interestingness

(objective interestingness) and business focused-interestingness

(subjective interestingness). Measuring actionability of knowledge is to

recognise statistically interesting patterns permitting users to react to

them to better service business objectives. The measurement of

knowledge actionability should be from perspectives of both objective

and subjective interestingness.

Definition 1. Actionability of a pattern: Given a pattern P, its

actionable capability act() is described as to what degree it can satisfy

both technical interestingness and business one.

If a pattern is automatically discovered by a data mining model if it

only satisfies technical interestingness request, it is usually called an

(technically) interesting pattern. It is presented as

∀ x ∈ I, ∃P : x.tech_int(P) ∧ x.act(P)

Page 8: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Actionable Knowledge Discovery: The Analytics Hierarchy Process Approach

92

In a special case, if either technical and business interestingness, or

hybrid interestingness measure integrating both aspects, are satisfied, it

is called an actionable pattern. It is not only interesting to data miners,

but generally interesting to decision-makers.

∀ x ∈ I, ∃P : x.tech_int(P) ∧ x.biz_int(P) ∧ x.act(P)

Therefore, the work of actionable knowledge discovery must focus

on knowledge findings, which can not only satisfy technical

interestingness but also business measures.

The real requirements for discovering actionable knowledge in

constraint-based context imply that real data mining is more likely to

be human involved rather than automated. Human involvement is

embodied through the cooperation between humans (including users

and business analysts, mainly domain experts) and data mining system.

This is achieved through the complementation between human

qualitative intelligence such as domain knowledge and field

supervision, and mining quantitative intelligence like computational

capability. Therefore, real-world data mining involves a human-

machine- collaborative knowledge discovery process.

3. Comparison Between Knowledge Discovery Process (KDP)

And Actionable Knowledge Discovery (AKD)

The basic differences between KDP and AKD are show in table 1.

These differences are described from deferent aspects as illustrated

below.

ASPECTS KDP AKD (DOMAIN – DRIVEN)

OBJECT

MINED

Data tells the story Data and Domain tells the story

AIM Develop innovative

approach

Generate business impacts

OBJECTIVE Algorithms are the focus Solving business problem is the

focus

Page 9: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Ikuvwerha L.O., Odumuyiwa, V.T, Ogunbiyi T.D, Uwadia, C.O., Abass O

93

DATA SET Mining abstract and refined

data set

Mining constraints real- life data

PROCESS Data mining is an

automated process

Humans are integrated into the

process

EVALUATION Based on technical metrics Based on actionable options

GOAL Let data create and verify

research innovation. Push

novel algorithms to

discover knowledge of

research interest.

Let data and metasynthetic

knowledge tell the hidden

business story. Discover

actionable knowledge to satisfy

end user

TABLE 1: COMPARISONS OF KDP AND AKD (DOMAIN-DRIVEN MINING).

4. The Analytic Hierarchy Process (AHP)

The foundation of the Analytic Hierarchy Process (AHP) is a set of

axioms that carefully delimits the scope of the problem environment

(Saaty 1996). It is based on the well-defined mathematical structure of

consistent matrices and their associated right- eigenvector's ability to

generate true or approximate weights, (Saaty,1980). The AHP

methodology compares criteria, or alternatives with respect to a

criterion, in a natural, pair wise mode. To do so, the AHP uses a

fundamental scale of absolute numbers that has been proven in practice

and validated by physical and decision problem experiments. The

fundamental scale has been shown to be a scale that captures individual

preferences with respect to quantitative and qualitative attributes just as

well or better than other scales (Saaty 1980). It converts individual

preferences into ratio scale weights that can be combined into a linear

additive weight w(a) for each alternative a. The resultant w(a) can be

used to compare and rank the alternatives and, hence, assist the decision

maker in making a choice. The basic steps in AHP are reasonable

descriptors of how an individual comes naturally to resolving a multi

criteria decision problem, then the AHP can be considered to be both a

descriptive and prescriptive model of decision making. The AHP is

perhaps, the most widely used decision making approach in the world

today. Its validity is based on the many hundreds (now thousands) of

Page 10: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Actionable Knowledge Discovery: The Analytics Hierarchy Process Approach

94

actual applications in which the AHP results were accepted and used by

the cognizant Decision Makers, Saaty (1996).

An example of such a hierarchy is presented in Figure 2. At the top

level, a goal is specified, in this case sustainable catchment use. At the

second level, all the objectives or criteria are listed, which in this

example are environmental, economic and social objectives. At the

bottom level, all the decision options are presented

Figure 2: Example of an AHP structure

Criteria or objectives can be divided into sub- or sub-sub-criteria

(objectives) for additional information and for clarification and

refinement. Criteria can be subjective (such as impact of trees on

recreational values) or objective (such as tree planting cost), depending

on the means used in evaluating the contribution of those criteria below

them in the hierarchy. Criteria are regarded as mutually exclusive and

do not depend on the elements below them in the hierarchy.

5. PROPOSED CONCEPTUAL MODEL.

The proposed model is called Analytic Hierarchy Process –

Actionable knowledge Discovery (AHP-AKD). The proposed

conceptual model is developed based on the six-step knowledge

Discovery process model (Pal, and Jain 2005). Apart from the first step,

the other five steps are grouped into three tiers which are Data tier, Data

Mining Tiers and Knowledge Discovery Tier. This is shown in Figure

4.

Page 11: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Ikuvwerha L.O., Odumuyiwa, V.T, Ogunbiyi T.D, Uwadia, C.O., Abass O

95

5.1. THE PROPOSED MODEL FRAMEWORK.

We considered the six-step knowledge discovery process to develop the

proposed model (Hybrid model). The description of the proposed model

is as follow:

STEP 1: Understanding the Problem Domain: it is at this step we define

the problems that needed to be solved.

STEP 2: AHP to formulate the problem structure: The problem

structure is formulated by employing AHP model.

STEP 3: Understanding of the data: this step involves the collection of

sample data and deciding which data is needed to achieve the project

goals and to solve the problem in step 1. Data are checked for missing

value, redundancy, completeness etc.

STEP 4: Preparation of the data: this is the step where we select the

data that is to be used for the Data mining methods or algorithms.

STEP 5: Data Mining: it is at this stage we apply various data mining

methods and tools. The methods or algorithms that will be used at this

Knowledge Discovery Tier

Data Mining Tier

Data Tier

Understanding of the

problem

Understanding of the

Data

Preparation of the

Data

DATA MINING

PROCESS

Evaluation of the

Discovered knowledge

Use of the

Knowledge

Discovered

Model :

is

knowle

dge

actiona

ble

Knowledge

patterns,

rules, clusters,

etc

Data

Source

Knowledge

Base

Extend Knowledge

to other Domain

N

o

Y

e

s

AHP TO

FORMULATE

DOMAIN

PROBLEM

Page 12: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Actionable Knowledge Discovery: The Analytics Hierarchy Process Approach

96

step are based on the identified project goals that are translated into data

mining goals in step 1.

Figure 4: THE PROPOSED CONCEPTUAL MODEL AHP-AKD.

STEP 6: Evaluation of the discovered Knowledge: the discovered

knowledge (either in form of patterns, rules, clusters etc) is checked to

see whether they are novel, interesting and actionable. Domain experts

are involved in the interpretation of the results.

STEP 7: Use of discovered Knowledge: this is the final step where the

discovered knowledge is used by the end user or decision maker to plan

and take action.

This proposed model has several feedback mechanisms. The entire

process is iterative. The expected iterative feedback loops are:

1. From step 3 to step1: This loop will help us to have a better

understanding between the data base and the problem domain.

2. From step 3 to step 2: A proper and further understanding of

the data will enable us to know the best preprocessing

algorithms to use.

3. From step 4 to step1: when the data mining process generates

unsatisfactory result, the process could go back to step1 to see

if the project goals should be modified or the problem is not

well understood.

4. From step 4 to step 3: this loop is necessary because some data

mining methods require us to prepare the data in such a way

that it could fit into the method. This is not known early in step

3 because we don’t know the data mining method or tool that

will be used.

5. From step 5 to step 1: The model we propose to develop in step

5 will check for the actionability, novelty, and interestingness

of the knowledge discovered. If the discovered knowledge is

not actionable, this could be as a result of incorrect

understanding of the problem domain, requirement and goals.

The process goes back to step 1.

Page 13: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Ikuvwerha L.O., Odumuyiwa, V.T, Ogunbiyi T.D, Uwadia, C.O., Abass O

97

6. Illustrative Example/Experimental Result

According to mcgarry (2005), a data mining algorithm produced the

following patterns.

Patterns 1: IF (age > 60) ∧ (salary = high) THEN

loan =approved

Patterns 2: IF (age < 60) ∧ (salary = average) ∧ (Record = poor)

THEN

loan = not approved

Patterns 3: IF (age < 60) ∧ (salary = low) THEN

loan = approved

While the end-user/expert defined pattern is

IF (age > 50) ∧(salary = low ) THEN

loan = not approved.

This patterns are the ones that have met with the objective

interestingness measure based on the data mining algorithms. We can

now use AHP to determine the actionability of the pattern produced.

We are proposing AHP as a subjective measure of interestingness.

This problem can be structured using AHP as follow

Fig 5: structured AHP for actionable pattern.

Novel

Actionable Pattern

Actionable Unexpected

Pattern 1 Pattern 3 Pattern 2

Page 14: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Actionable Knowledge Discovery: The Analytics Hierarchy Process Approach

98

The goal is to find actionable pattern. The criteria used are

actionability, unexpected and novel. The alternatives are Pattern 1,

pattern 2 and pattern 3.

6.1. Pairwise Comparison Judgment Matrices of Model

The pair-wise comparison matrix from all the expert is used to

calculate the criterion weight. A square matrix is used to represent all

such judgments in making comparisons with respect to a single factor

by comparing the set of elements. Each judgment represents the

dominance of an element in the left column of the matrix over an

element in the row on top. In making the final decision, all judgments

with respect to some factors or property to be processed were

synthesized along with other matrices of comparisons involved. The

values of consistency ratio (CR) lower than 0.10 define the matrix A as

acceptable, the values that are slightly higher (between 0.10 and 0.20)

must be considered with care, matrix A should be rejected with higher

values (greater than 0.20).

PAIRWISE MATRIX RESULTS

Table 1: Pairwise comparison matrix with respect to the factors

λmax = 3.041 CR= 0.0356

This result shows that actionable is of more important with 57%,

followed by unexpectedness with 32%, and Novel with 11%. In finding

actionable patterns or knowledge, actionability of the pattern comes

first followed by unexpectedness and novel.

Table 2: Pairwise comparison matrix for the Alternative with

respect actionable factor

FACTORS ACTIONABLE

UNEXPECTED

NOVEL

NORMALISED

EIGEN

VECTOR

ACTIONABLE

1 2 5 0.5701

UNEXPECTED

1/2 1 3 0.3207

NOVEL

1/5 1/3 1 0.1092

Page 15: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Ikuvwerha L.O., Odumuyiwa, V.T, Ogunbiyi T.D, Uwadia, C.O., Abass O

99

λmax = 3.002 CR= 0.0138

the patterns are evaluated according to their actionability. We find

out that pattern 1 is more actionable with 65%, followed by Pattern 2

with 23%, and pattern 3 with 12%.

Table 3: Pairwise comparison matrix for the Alternative with

respect unexpected factor

λmax = 3.0392 CR= 0.033.

This result shows that according to pattern unexpectedness, the

pattern are ranked as follow: pattern 3 with 64.13%, pattern 2 with

23.755 and Pattern 1 with 12.14%. From this it is clear that pattern 3

contradicts the user’s belief and it is therefore unexpected. This also

confirm the result from the Mcgarry (2005) results using

unexpectedness as a factor.

Table 4: Pairwise comparison matrix for the Alternative with

respect Novel factor

ACTIONABL

E

PATTER

N 1

PATTER

N 2

PATTER

N 3

NORMALISE

D EIGEN

VECTOR PATTERN 1

1 5 3 0.6485

PATTERN 2

0.33 1 3 0.2296

PATTERN 3

0.2 0.33 1 0.1219

UNEXPECTED

PATTERN 1

PATTERN 2

PATTERN

3

NORMALISED

EIGEN

VECTOR PATTERN 1

1 0.5 0.2 0.1213

PATTERN 2

2 1 0.33 0.2374

PATTERN 3

5 3 1 0.6413

Page 16: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Actionable Knowledge Discovery: The Analytics Hierarchy Process Approach

100

λmax = 3.038 CR= 0.033

The three factors results are integrated together to form the overall

priority table which is shown in table 5.

Table 5: overall priority

ALTERNATIVES OVREALL PRIORITY

PATTERN 1 0.4664

PATTERN 2 0.2407

PATTERN 3 0.2929

This result shows that the pattern are ranked as follow: pattern 1 with

46.64%, pattern 2 with 24.07% and Pattern 3 with 29.29%. From this it

is clear that pattern 1 is seen to be more actionable followed by pattern

3 and then pattern 2

7. Conclusion

Much of the research in the area of Knowledge Discovery in

Databases (KDD) has focused on the development of more efficient and

effective data mining algorithms. However, recently, issues related to

the usability of these techniques in extracting exploitable knowledge

from databases has drawn significant attention. Real-world data mining

applications for knowledge discovery have proposed urgent requests

NOVEL

PATTERN

1

UNEXPECTED

PATTERN 2

PATTERN

3

NORMALISED EIGEN

VECTOR

PATTERN

1

1 2 3 0.5295

PATTERN

2

0.33 1 2 0.3088

PATTERN

3

0.33 0.5 1 0.1617

Page 17: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Ikuvwerha L.O., Odumuyiwa, V.T, Ogunbiyi T.D, Uwadia, C.O., Abass O

101

for discovering actionable knowledge of main interest to real user and

business needs. The major issue in actionable knowledge discovery is

the interestingness measure: objective and subjective measure. The

proposed conceptual model uses the AHP as the subjective measure and

the data mining Algorithms as the objective measure. It is clear from

table 5 that Pattern 1 is the most actionable pattern based on the end

user’s. It has an average priority of 47% . Pattern 3 is next with an

average priority of 29%. This is then followed by pattern 2 with an

average priority of 24%. This research therefore concludes that AHP

can be effectively used as subjective interestingness measure for

actionable knowledge.

List of references

Amruta, L. & Balachandra, k. (2013) Performance Evaluation of

Actionable Knowledge Discovery (AKD) Framework under the

Decision Making System. International journal of advanced

research in computer science and engineering. 3(9).

Ankerst, M., (2002): Report on the SIGKDD-2002 Panel the Perfect

Data Mining Tool: Interactive or Automated? ACM SIGKDD

Explorations Newsletter, 4(2):110-111.

Cao, L., Zhang, C.(2007a): Knowledge Actionability: Satisfying

Technical and Business Interestingness International Journal of

Business Intelligence and Data Mining, 2(4): 496-514

Cao, L., Zhang, C.(2007b): The Evolution of KDD: Towards Domain-

Driven Data Mining, International Journal of Pattern Recognition

and Artificial Intelligence, 21(4): 677-692

Cao, L., (2007). Domain-Driven Actionable Knowledge Discovery,

IEEE Intelligent Systems, 22(4):78-89.

Cao, L., (2012). Actionable Knowledge Discovery and delivery,

WIREs Data mining Knowledge discovery 2012, 2:149-163

Cao, L., & Zhang, C. (2008). Domain-driven Data Mining.

PAKDD2006 (pp. 821-830). LNAI 3918.

Dharanik & Gudikandula, K. (2012). Actionable knowledge discovery

using multi-step mining. International journal of computer science

and network (IJCSN). Vol. 1 issue 6

Page 18: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Actionable Knowledge Discovery: The Analytics Hierarchy Process Approach

102

Fayyad, U., Shapiro, G., Smyth, P.(1996): From Data Mining to

Knowledge Discovery in Databases, AI Magazine, 37-54,.

Fayyad, U., Shapiro, G., Uthurusamy, R.(2003): Summary from the

KDD-03 Panel - Data mining: The Next 10 Years, ACM SIGKDD

Explorations Newsletter, 5(2): 191-196,.

Goan, T.,(2000)“From Data to Actionable Knowledge: Applying Data

Mining to the Problem of Intrusion Detection”. International

Conference on Artificial Intelligence (IC-AI'2000)

Kavitha, K. & Ramaraj, E. (2013). An adequate approach for actionable

pattern using combine and composite association rule mining.

UNIASCIT, vol. 3 (2).

McGarry, K.(2005). A survey of interestingness measure for

knowledge discovery. The knowledge engineering review.

Cambridge university press.

Pal, N. R.,& Jain, L.C.(2005). Advanved Techniques in Knowledge

Discovery and Data Mining, Springer Verlag.

Piatetsky – Shapiro, G.,(1991). Knowledge Discovery in Real

Databases: A Report on the IJCAI-89 Workshop. AI Magazine

11(5): 68–70.

Saaty, T. L. (1996) Decision Making for Leaders: The Analytical

Hierarchy Process for

Decisions in a Complex World, The Analytical Hierarchy Process

Series, Vol. 2, pp 71-74.

Saaty, T.L. (1980) The Analytical Hierarchy Process, McGraw Hill,

N.Y.

Saaty, T.L. (2008) ‘Decision making with the Analytical Hierarchy

Process’, int. Journal service Sciences, Vol.1, No. 1, pp 83-98.

Page 19: Arts Science Faithtoki-ng.net/toki2016v2/sites/default/files/toki2016_papers/Actionable... · Analytic Hierarchy Process (AHP) approach is used to measure the subjective interestingness

Transition from Observation to Knowledge to Intelligence (TOKI) Conference is a forum that allows researchers from the fields of Competitive Intelligence, Internet of Things (IoT), Cloud Computing, Big Data and Territorial Intelligence to present their novel research findings and results. Common to all these fields are the concepts of information, information systems, knowledge, intelligence, decision-support systems, ubiquities, etc. The relevance of research findings, results obtained, systems developed and techniques adopted in these research fields for both the industries and government cannot be overemphasized.

Therefore, the Conference welcomes contributions in the following areas:

Smart Cities: With focus on Intelligent Transportation Systems, Observatory Systems, Smart Electricity Grids, building automation, assisted living and e-health management systems. Areas such as application of Geographical Information Systems, Territorial Intelligence and Sensors are also considered.

Big Data Analytics: This includes Big Data, Information Visualization, Data Analysis and related applications.

Semantic Web: Standardized formats and exchange protocols for web based data.

IoT Analytics: These center around innovative algorithms and data analysis techniques for extracting meaning and value from the Internet of Things.

Resource Management: This includes energy saving techniques, effective and efficient utilization of resources, intelligent data processing, mining, fusion, storage, and management, context awareness and ambient intelligence.

IoT Enabling Technologies: These center around technologies that drive pervasive / ubiquitous systems some of which include but not limited to IPv6, NFC, RFID and Microprocessors.

Interoperable and Adaptive Information Systems: These include but are not limited to Decision Support Systems, collaborative and co-operative systems and other forms of systems that support interfacing of multiple elements and entities.

Mobile IoT: Smart phone applications for generating and consuming data, crowd sourced data, e-commerce, mobile advertising, B2B, B2C and C2C connectedness.

Cloud Computing: Including security, storage and access to data stored in the cloud;

service provisioning and resource utilization; cloud communication protocols;

interoperability among users and devices with respect to linked data.

Editors Prof. Amos DAVID & Prof. Charles UWADIA

9782954676036