=Paper=
{{Paper
|id=Vol-2158/paper7
|storemode=property
|title=Significance Level of a Big Data Query by Exploiting Business Processes and Strategies
|pdfUrl=https://ceur-ws.org/Vol-2158/paper7.pdf
|volume=Vol-2158
|authors=Loan Thi Ngoc Dinh,Gour Karmakar,Joarder Kamruzzaman,Andrew Stranieri
|dblpUrl=https://dblp.org/rec/conf/balt/DinhKKS18
}}
==Significance Level of a Big Data Query by Exploiting Business Processes and Strategies==
https://ceur-ws.org/Vol-2158/paper7.pdf
Significance Level of a Big Data Query by Exploiting
Business Processes and Strategies
Loan Thi Ngoc Dinh, Gour Karmakar, Joarder Kamruzzaman, and Andrew Stranieri
Federation University Australia
{l.dinh; gour.karmakar; joarder.kamruzzaman;
a.stranieri}@federation.edu.au
Abstract. Querying data is one of the most frequent activities in business or-
ganisations. The tasks involving queries for big data collection, extraction and
analysis have never been easy, because to obtain the high quality responses, the
expected outcome from these tasks need to be more accurate and highly rele-
vant to a business organisation. The emergence of big data era has further com-
plicated the task. The enormous volume of data from diverse sources and the
variety of queries impose a big challenge on business organisations on how to
extract deep insight from big data within acceptable time. Determining signifi-
cance levels of queries based on their relevance to business organisations is able
to deal with such challenge. To address this issue, up to our knowledge, there
exists only one approach in the literature to calculate the significance level of a
query. However, in this approach, only business processes are considered by
manually selecting weights for core and non-core business processes. As the
significance level of a query must express the importance of that query to a
business organisation, it has to be calculated based on the consideration of
business strategic direction, which requires the consideration of both business
processes and strategies. This paper proposes an approach for the first time
where the significance level of a query is determined by exploiting process con-
tributions and strategy priorities. The results produced by our proposed ap-
proach using a business case study show the queries that are associated with
more important business processes and higher priority strategies have higher
significance levels. This vindicates the application of the significance level in a
query to dynamically scale the semantic information use in capturing the ap-
propriate level of deep insight and relevant information required for a business
organisation.
Keywords: Semantic similarity, Significance level, Business process, Business
strategy, Query processing
1 Introduction
Under the influence of big data, query analysis has shown a huge difficulty for busi-
ness organisations to handle their data processing systems [1-3]. To deal with such a
challenge, a number of approaches and applications have been introduced. Some of
Lupeikiene A., Matulevičius R., Vasilecas O. (eds.):
Baltic DB&IS 2018 Joint Proceedings of the Conference Forum and Doctoral Consortium.
Copyright © 2018 for this paper by the papers' authors. Copying permitted for private and academic purposes.
them have focused on how to increase the speed of data collection and analytics. The
others are concentrating on the techniques to improve the accuracy of data processing
and analysis.
To improve the speed of data analytics, several big data processing platforms and
tools have already been introduced such as Hadoop, Spark, Kafka and Tableau (tab-
leau.com). There also exist a number of studies to reduce redundant, irrelevant data
during data collection and analysis for queries.
Nanchani et al. [4] introduced a solution for avoidance of duplication in storing
customer records. Whereas, [5] and [6] presented the solutions in order to deal with
redundant data. Besides that, since big data has led to the outcome delays and infor-
mation overload, the techniques for big data collection and analytics have also been
improved such as real-time data collection in [7, 8] and speed optimization during
data processing in [4, 9]. Additionally, search engines (e.g., Google Trends and
Baidu) have proven advantages in extracting useful business information for organisa-
tions [10, 11].
In addition to analytics, retrieving the deep insight of data also requires exhaustive
data collection from all possible sources. Even with exhaustive data collection, a big
data query cannot obtain deep insight from all relevant unstructured big data without
semantic interpretation in both data collection and processing. This is because differ-
ent big data sources can represent the same semantic information in many different
ways. A huge number of data sources continuously generating an enormous amount
of data and the requirement of semantic interpretation have also made big data collec-
tion and processing computationally expensive. To address these issues associated
with obtaining deep insight and computational complexity, we need to dynamically
scale the use of semantic interpretation in both data collection and processing and data
sources based on the importance of a query.
The approach introduced in [1] can determine the significance level of a query.
This approach claims that by considering the significance level of a query, the amount
of time spent on less important queries can be reduced. For capturing deep insight, the
data processing systems can focus more on the queries that have higher significance
level. Nevertheless, in this approach, only business processes are considered by
manually selecting weights for core and non-core business processes. Besides this,
since the business strategies are not utilised at all, the significance level of a query
calculated by this approach does not fully cover the objectives of a business organisa-
tion and hence cannot capture the deep insight of data relevant to fulfil the goals of an
organisation.
As none of studies has worked on calculating significance level of a query based
on the aspect of business strategy, thereby for the first time, we aim to introduce a
method that determines the significance level of a query reflecting the business objec-
tives. The contributions in this paper are highlighted as below:
• We determine the significance level of a query considering both business proc-
esses and organization strategies.
• The business process contributions are calculated based on the extent of the con-
tribution of a process to business strategies, the number of strategies in which a proc-
ess contributes to and the priority score of that strategy toward business organisation.
64
• The results produced by the proposed approach from our business case study
have shown that the queries that are related to more important business processes and
higher priority strategies, have higher significance levels.
The structure of this paper is as follows: Our proposed methodology is presented in
Section 2. The results of the case study in a scenario of retail enterprise are described
in Section 3. Section 4 concludes the paper.
2 Proposed Approach to Determine the Significance Level of a Query
As alluded before, a significance level of a query needs to reflect the importance of
that query for a business organisation. Queries are normally very short and a business
organisation is a combination of many activities and transactions. These raise an im-
portant question ’How to find a link between a query and a business organisation’. To
answer this question, we have decided to choose business process and strategy as the
main representation for a business organisation. This is because business processes
and strategies are the two key entities that always play the main role in a business
organisation [14]. Business processes represent all the activities and transactions in
business organisations. Business strategies play vital roles in business development
including earning more revenue, achieving strategic advantage and its expansion. A
significance level of a query is thereby determined based on the relationship between
query, process and strategy as seen in Fig. 1.
Fig. 1. The overall process of proposed approach to determine the significance level of a query
To our knowledge, the only approach available in the literature was introduced in
[1]. As mentioned before, this approach determines the significance level of a query
by considering only business processes. Moreover, the contribution of a process is
intuitively assigned as two values, i.e., 0.5 or 1. The value 1 is for core business proc-
esses, while 0.5 for non-core business processes. Dividing processes into two groups
(i) core and (ii) non-core does not fully reflect the strategic direction of a business
organisation. Because some processes may have either very high or very low effect on
a business organisation, while the others may have an effect that are fluctuating be-
65
tween the two extremes. This indicates that the contribution of a process should be a
continuous value in [0, 1]. On the other hand, the business strategies that describe
business tactics to achieve the business objectives and goals have not been taken into
account in determining the significance level of query. Therefore, this demands the
introduction of a new approach that is able to calculate the significance level of a
query reflecting a process contribution more accurately and integrating the business
strategies.
To embed the more accurate reflection of a business objective in a big data query,
we aim to introduce an approach to calculate the significance level of a query based
on the contribution weight of a process and the priority level of a strategy. Note, a
contribution weight refers to how much impact a process p has on the strategy satis-
faction [15]. For example, Table 1 shows a strategy S3 has three contributing proc-
esses P2, P4, P5, P6 and P12, whose contributions are 90%, 70%, 70%, 50% and
80%, respectively.
Table 1. Contribution weight of processes
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14
S1 90% 90% 20% 90% 90% 50% 0% 0% 50% 65% 10% 65% 55% 0%
S2 50% 90% 45% 60% 90% 90% 70% 55% 50% 70% 20% 70% 70% 55%
S3 0% 90% 0% 70% 70% 50% 0% 0% 0% 0% 0% 80% 0% 0%
S4 0% 90% 0% 60% 70% 50% 0% 0% 0% 0% 0% 80% 0% 0%
S5 0% 90% 80% 75% 75% 80% 0% 0% 0% 60% 0% 80% 70% 0%
S6 0% 90% 90% 85% 85% 90% 0% 0% 0% 0% 80% 80% 70% 0%
S7 0% 90% 0% 70% 70% 50% 0% 0% 0% 0% 0% 80% 0% 0%
S8 0% 90% 0% 85% 90% 75% 0% 0% 0% 75% 0% 85% 90% 0%
S9 0% 90% 35% 10% 10% 10% 70% 50% 50% 65% 65% 0% 85% 40%
S10 0% 90% 70% 65% 65% 20% 85% 75% 80% 75% 75% 0% 75% 0%
S11 0% 90% 10% 50% 50% 75% 70% 50% 60% 70% 75% 0% 85% 0%
S12 0% 90% 10% 65% 65% 45% 0% 0% 0% 75% 65% 0% 70% 0%
S13 0% 90% 40% 85% 85% 65% 70% 40% 70% 80% 70% 0% 75% 0%
S14 0% 90% 70% 90% 90% 35% 70% 50% 0% 80% 75% 0% 80% 50%
S15 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 90%
S16 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 90%
S17 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 90%
S18 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 90%
S19 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 80%
S20 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 80%
S21 0% 90% 90% 50% 50% 55% 0% 0% 0% 0% 55% 0% 0% 0%
S22 70% 90% 75% 70% 90% 70% 0% 0% 0% 80% 55% 80% 85% 0%
S23 0% 90% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 90%
The term "strategy priority" refers to the task of ranking strategic objectives based
on their importance in business organisation. By clarifying their roles and urgency, a
business organisation has a better view of what to do first, what need to be focused
more than the others [16, 17].
To consider process contributions to strategies and a strategy priority level in de-
termining the significance level of a query, firstly, we need to find the process-
strategy relationship. This can be done using a model introduced in [18] where the
link between a strategy and a process is built by adopting a rule-based inference
model.
Besides the importance of realising which strategies a process contributes to, sec-
ondly, we then need to indicate how successful a strategy will have when its related
66
process is completely executed. This is called a contribution weight of a process for a
specific strategy [15]. These contribution weights can be determined by the chief
information officer or business analyst or the manager of a business organisation.
Thirdly, we need to consider how to determine the strategy priorities. A strategy
priority represents the extent of importance and urgency of a strategy. This means the
more importance and urgency, the higher priority the strategy has. By clarifying the
strategy priority, a business organisation has a better view of what to do first and what
needs to be focused on [16, 17].
In this approach, to determine the significance level of a query, five main values
depicted in Fig. 1 need to be calculated: (i) the similarity scores between process p
and query q, (Block 1 of Figure 1), (ii) the number of strategies in which proc-
ess p contributes to (Block 2), (iii) the contribution of process p to strategy s,
(Block 3), (iv) the strategy priority of strategy s, (Block 4), and (v) the final con-
tribution of process p, using , and (Block 5). Note, as described above,
Block 5 represents the function for the calculation of final contribution process . In
Fig. 1, this block also shows that the calculation of requires the execution of the
functions represented by Blocks 2, 3 and 4.
The concept of semantic information is also valuable to model the business proc-
esses. The business processes that are semantically annotated have shown the enrich-
ment in their process descriptions. A set of carefully selected semantic annotations
over a business process can not only reduce the ambiguity to represent the logically
connected activities of that business process for fulfilling a certain business goal but
also provide business analysts with a better understanding about the business proc-
esses. Additionally, semantic annotations can also make a relationship between the
business processes and the other characteristics of an organisation. Thus, the semantic
annotation scheme allows us to link a query, business processes and business strate-
gies semantically by avoiding mismatched and unstructured knowledge representation
involved in a business process model [12, 13]. This has motivated us to calculate a
similarity score between a query and a business process by exploiting the semantic
similarity scores between the keywords of a query and the annotations of a business
process.
Suppose query q consists of n keywords {k1, k2, …, kn} and process p contains m
annotations {a1, a2, …, am}. The semantic similarity score between a process and
a query is calculated as shown below:
where,
Here, h is the depth of the sub-sumer of ki and aj in the hierarchical semantic tree
of WordNet (https://wordnet.princeton.edu/). The value l is the shortest path between
ki and aj in the hierarchical semantic tree of WordNet and this is calculated as men-
tioned in [19].
67
As mentioned before, the contribution of process p to strategy s, can be as-
signed by chief information officer or business analyst. The strategy priority of
strategy s can be determined based on a business strategy prioritization tool [20].
We can then calculate the final contribution of a process is as follows:
(i) The contribution of a process should be determined by using a utility function
that follows the diminishing property of microeconomic model [21], i.e., the contribu-
tion of a process should follow a logarithmic function. This means that similar to
logarithmic curve, it should increase with increasing number of associated strategies
up to a certain limit and after that the increment in contribution should slow down.
Therefore, we can define the contribution of process p, as,
where, z is the total number of strategies considered, 0≤ ≤1 for yp≤z and c repre-
sents the sensitivity of the process contribution with respect to the number of associ-
ated strategies. The process contribution ( ) derived in (2) was plotted over the
number of associated strategies (yp) for different values of c=0.2, 0.5, 1.0, 2.0 and
z=23, and is shown in Figure 2. Figure 2 shows for a constant value of z, the higher
the value of c, the higher sensitivity of the process contribution with the number of
business strategies in which a process contributes is. Therefore, for a particular busi-
ness process, if it requires to quickly vary the process contribution over the number of
associated strategies, the higher value of c needs to be used.
Fig. 2. The process contribution vs the number of associated strategies for different values of
c=0.2, 0.5, 1.0, 2.0 and z=23
(ii) Considering the contribution of a process to all strategies, and the strategies’
priority, we can define the following aggregation function to calculate the contribu-
tion of a process p, using and as,
68
Eq. (3) exhibits a consensus view that the higher the values of and , the
higher value of is.
Assuming the contribution of a process to the extent of covering the number of
strategies, and the amount of contribution to all strategies and their priority levels as
equally important, we can formulate an equation to determine the final contribution
of process p in terms of a combination of and defined in (2) and (3), respec-
tively in the following way,
where, max( ) is the maximum value of all and used to normalise the value in
[0, 1].
Finally, similar to (2), assuming the significance level of a query varies non-
linearly following a logarithmic curve with its similarity score with a process and the
process contributions, the significance level of a query q can be determined using the
similarity scored defined in (1) and the final contribution derived in (4) as,
with t is the total number of business processes.
3 Case Study
The proposed approach was implemented using Pyke package in Python program-
ming language for establishing the relationship between a process and a strategy. The
contribution weights of the processes were intuitively assigned and the strategy priori-
ties shown in Table 2 were determined by using the strategy prioritization tool [20].
This case study consists of 18 queries - Q1 to Q18 as shown in [1], 14 business proc-
esses - P1 to P14 and 23 business strategies - S1 to S23 as described in [18].
Table 2. Priority score of strategies
Strategy Priority Strategy Priority Strategy Priority
S1 7.1 S9 4.5 S17 3.5
S2 7.7 S10 7.8 S18 2.8
S3 4.8 S11 4.9 S19 5.3
S4 5.4 S12 4.8 S20 3.1
S5 7.1 S13 6.9 S21 5.7
S6 7.4 S14 5.4 S22 6.1
S7 5.0 S15 3.9 S23 4.8
S8 5.4 S16 3.2
The strategy priorities were calculated using a software tool [20] and the priority
value for each strategy are presented in Table 2. According to this tool, the strategy
69
priority is determined by following three main criteria - (i) strategic fit, (ii) economic
impact and (iii) feasibility. The elements of each criteria were weighted depending on
their importance. The elements of each criteria for a specific strategy were then
ranked between 1 to 10. The more important an element is, the higher rank it has. As
a representative example, the calculated rank of each element of strategy S1 including
S1’s priority value (7.1) is presented in Table 3. Finally, the calculated priority value
of each strategy is listed in Table 2. The results show S2 having priority value 7.7 is
the highest priority, while S18 with priority value 2.8 is the lowest.
Table 3. An example for strategy priority of S1
Criteria Element Weight Rank for S1
Strategic fit Alignment with company goals 15% 9
Market positioning 15% 9
Core capabilities 10% 7
Economic impact Revenue potential 10% 7
Profitability & margin 15% 6
Growth potential 15% 8
Feasibility Technical risk 10% 1
Resources - Financial 5% 8
Resources - People 5% 7
Strategy priority for S1 7.1
The similarity score between each query and each process was calculated using
(1). These scores are as shown in Table 4 that describes the highest (0.9994) similar-
ity scores were obtained between Q1 and four processes {P3, P6, P7, P8} and the
lowest (0.1505) similarity score was between Q8 and P9.
According to a rule-based inference model introduced in [18], we have applied this
model to our scenario and found that the process P1 is associated with the strategies
S1, S2 and S22. Similarly, the relationship of other processes with their relevant
strategies was discovered using the same knowledge base. Then the contribution
weights of all processes for each strategy were intuitively assigned and are given as
described in Table 1. For example, the process P1 has the highest contribution weight
on the strategy S1. This means after P1 is completely executed, the strategy S1 is
expected 90% chance of success. For those places where the contribution weights are
0% such as the contribution weights of P1 on the strategies S3 to S21 and S23, this
means there is no rule in knowledge base that reflects the link between P1 and these
strategies or there is no effect of P1 on these strategies. Another example, strategy S7
has five contributing processes P2, P4, P5, P6 and P12, whose contributions are 90%,
70%, 70%, 50% and 80%, respectively. This means if the process P2 is completely
executed, there is a 90% chance that strategy S7 is achieved. Similarly, if P4, P5, P6
or P12 is completely executed, there is a 70%, 70%, 50% or 80% of chance for the
strategy S7 to be achieved.
70
Table 4. Similarity scores between queries and processes
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14
Q1 0.86 0.26 1.00 0.20 0.74 1.00 1.00 1.00 0.19 0.38 0.32 0.30 0.17 0.45
Q2 0.34 0.69 0.46 0.57 0.57 0.48 0.55 0.55 0.29 0.56 0.53 0.61 0.56 0.43
Q3 0.29 0.42 0.28 0.73 0.50 0.33 0.46 0.46 0.31 0.46 0.42 0.56 0.40 0.60
Q4 0.25 0.41 0.34 0.49 0.45 0.29 0.44 0.44 0.17 0.41 0.30 0.45 0.37 0.51
Q5 0.42 0.56 0.39 0.56 0.56 0.43 0.47 0.45 0.18 0.56 0.39 0.56 0.47 0.51
Q6 0.60 0.39 0.46 0.40 0.57 0.56 0.43 0.43 0.25 0.37 0.41 0.40 0.29 0.57
Q7 0.43 0.41 0.43 0.47 0.62 0.39 0.56 0.56 0.22 0.42 0.28 0.49 0.40 0.38
Q8 0.36 0.49 0.41 0.43 0.54 0.35 0.52 0.52 0.15 0.67 0.21 0.42 0.40 0.38
Q9 0.36 0.44 0.44 0.44 0.52 0.37 0.71 0.62 0.26 0.44 0.32 0.40 0.38 0.36
Q10 0.52 0.45 0.47 0.43 0.64 0.49 0.59 0.59 0.28 0.44 0.28 0.45 0.41 0.34
Q11 0.45 0.42 0.41 0.46 0.62 0.46 0.54 0.54 0.31 0.41 0.30 0.48 0.37 0.42
Q12 0.40 0.42 0.46 0.40 0.63 0.45 0.57 0.57 0.22 0.45 0.29 0.43 0.42 0.28
Q13 0.47 0.45 0.44 0.50 0.53 0.37 0.39 0.39 0.23 0.53 0.41 0.38 0.41 0.70
Q14 0.25 0.34 0.28 0.37 0.31 0.21 0.34 0.28 0.23 0.38 0.36 0.42 0.28 0.40
Q15 0.39 0.49 0.26 0.42 0.46 0.35 0.53 0.50 0.21 0.40 0.35 0.45 0.30 0.46
Q16 0.21 0.29 0.24 0.34 0.45 0.15 0.29 0.29 0.21 0.32 0.28 0.29 0.29 0.29
Q17 0.37 0.27 0.37 0.36 0.42 0.28 0.33 0.33 0.23 0.37 0.26 0.38 0.27 0.45
Q18 0.32 0.50 0.25 0.43 0.39 0.38 0.30 0.28 0.24 0.33 0.36 0.48 0.27 0.49
Fig. 3 shows the significance level of all queries produced by our proposed ap-
proach using (5). To show the impact of c used in (2) on the significance level of all
queries, we used three different values of c such as 0.5, 1 and 2 keeping the value of
other parameters as the same. Figure 3 shows the lower the value of c, the higher the
significance level of all queries is. This is because for a constant number of total
strategies (e.g., 23 strategies have been used in this case study), the lower the value of
c, the lower the effect of the number of strategies associated with a business process
(yp) is. Therefore, c provides a way to make a trade-off between and to a
certain extent. For the sake of clarity, in the remainder of this paper, we will discuss
the results only for c = 1. The significance levels with c = 1 are between 0.384 and
0.547. The results show that the highest significance level is 0.547. This is because,
query Q2 has the highest similarity score 0.6864 with process P2. The contribution
weights of process P2 with strategies S1 to S14 and S21 to S23 are 90%. The priority
of strategies S1, S2, S5, S6, S10 and S13 are in the top of highest priority scores.
Fig. 3. Significance levels of queries
71
On the other hand, query Q16 has the lowest significance level as 0.384. This is
mainly because this query has the highest similarity score 0.4519 with P5. Although,
P5 also has contribution with similar group of strategies like P2 except S23 but the
contribution weights of P5 are much lower than those of P2.
However, the significance level of the same queries computed by [1] without con-
sidering the contribution of a process to each strategy and the strategy priority is in
[0.570, 0.806]. If we compare this range values with the values [0.384, 0.547] that are
determined by our proposed approach for the same queries using the same processes,
it shows that process contribution and strategy priority have a significant impact on
accurately determining the significance level of queries.
In addition, in [1], the significance level of a query was calculated based on the
contribution weight of a process which was only defined by two values, i.e., 0.5 (non-
core) and 1 (core). Whereas, the contribution weight of the process in our proposed
approach is not narrowed within those two values, it is a continuous value within [0,
1]. This captures the process contribution more accurately in our proposed approach
than [1]. Furthermore, our proposed approach also exploits the contribution of a proc-
ess to strategies and the strategy priority directly. Therefore, it is expected that the
significance level of a query reflects the importance of a query for a business organi-
sation more accurately.
4 Conclusions
In this paper, a method for calculating the significance level of a query has been in-
troduced. The significance levels have reflected not only the importance of a query
toward business organisation but also emphasised the relevance between queries and
business strategies. The results from the implementation have demonstrated that, the
more relevance a query has, the higher score it owns. Based on this, the enterprise
information system in business organisation can understand which query is more im-
portant to focus and spend more time to extract deep insight of data. Therefore, this
can improve the efficiency and effectiveness of the information system in order to
manage big data better.
References
1. L. T. N. Dinh, G. Karmakar, J. Kamruzzaman, and A. Stranieri, “Significance level of a
query for enterprise data,” in 30th International Conference of Business Information Man-
agement Association (IBIMA), 2017.
2. D. G. Andersen and N. Feamster, “Challenges and Opportunities in Internet Data Mining,”
Parallel Data Laboratory, Carnegie Mellon University, Research Report CMU-PDL-06-
102, 2006.
3. M. H. ur Rehman, C. S. Liew, A. Abbas, P. P. Jayaraman, T. Y. Wah, and S. U. Khan,
“Big data reduction methods: a survey,” Data Science and Engineering, vol. 1, no. 4, pp.
265-284, 2016.
4. P. Nachnani, D. Milbrath, S. Levine, and E. Levine, “Method and system for cleaning data
in a customer relationship management system,” Nov. 15 2016, US Patent 9,495,403.
72
5. T.-H. Chen, W. Shang, Z. M. Jiang, A. E. Hassan, M. Nasser, and P. Flora, “Finding and
evaluating the performance impact of redundant data access for applications that are de-
veloped using object-relational mapping frameworks,” IEEE Transactions on Software
Engineering, vol. 42, no. 12, pp. 1148-1161, 2016.
6. M. H. ur Rehman, V. Chang, A. Batool, and T. Y. Wah, “Big data reduction framework
for value creation in sustainable enterprises,” International Journal of Information Man-
agement, vol. 36, no. 6, pp. 917-928, 2016.
7. M. R. Fellenz and M. Brady, “RFID and data capture technologies in global service supply
chains: meeting the information management challenge,” Communications of the IBIMA,
vol. 4, pp. 41-49, 2008.
8. P. Lubell-Doughtie, P. Pokharel, M. Johnston, and V. Modi, “Improving data collection
and monitoring through real-time data analysis,” in 3rd ACM Symposium on Computing
for Development, p. 28, 2013.
9. M. Jarke, M. Lenzerini, Y. Vassiliou, and P. Vassiliadis, “Query processing and optimiza-
tion,” in Fundamentals of Data Warehouses, Springer, pp. 107-122, 2003
10. L. Vaughan, “Discovering business information from search engine query data,” Online
Information Review, vol. 38, no. 4, pp. 562-574, 2014.
11. L. Vaughan and Y. Chen, “Data mining from web search queries: A comparison of Google
trends and baidu index,” Journal of the Association for Information Science and Technol-
ogy, vol. 66, no. 1, pp. 13-22, 2015.
12. D.M. Riehle, S. Jannaber, P. Delfmann, O. Thomas, and J. Becker, “Automatically Anno-
tating Business Process Models with Ontology Concepts at Design-Time,” in International
Conference on Conceptual Modeling, Springer, pp. 177-186, 2017.
13. C. Di Francescomarino, M. Rospocher, C. Ghidini, and A. Valerio, “The role of semantic
annotations in business process modelling,” In IEEE International Conference on Enter-
prise Distributed Object Computing (EDOC), pp. 181-189, 2014.
14. E. D. Morrison, A. K. Ghose, H. K. Dam, K. G. Hinge, and K. Hoesch-Klohe, “Strategic
alignment of business processes,” in International Conference on Service-Oriented Com-
puting, Springer, pp. 9-21, 2011.
15. R. Guizzardi and A. N. Reis, “A method to align goals and business processes,” in Interna-
tional Conference on Conceptual Modeling, Springer, pp. 79-93, 2015.
16. A. Nieto-Rodriguez. (2016) How to prioritize your company’s projects. [Online]. Avail-
able: https://hbr.org/2016/12/how-to-prioritizeyour-companys-projects
17. D. Lidow. (2017) A better way to set strategic priorities. [Online]. Available:
https://hbr.org/2017/02/a-better-way-to-set-strategic-priorities
18. L. T. N. Dinh, G. Karmakar, J. Kamruzzaman, and A. Stranieri, “A rule based inference
model to establish strategy-process relationship,” in 30th International Conference of
Business Information Management Association (IBIMA), 2017.
19. Y. Li, D. McLean, Z. A. Bandar, J. D. O’shea, and K. Crockett, “Sentence similarity based
on semantic nets and corpus statistics,” IEEE transactions on knowledge and data engi-
neering, vol. 18, no. 8, pp. 1138-1150, 2006.
20. High performance marketing – demand metric. Demand Metric. [Online]. Available:
https://www.demandmetric.com/content/businessstrategy-prioritization-tool
21. D. Niyato and E. Hossain, “A microeconomic model for hierarchical bandwidth sharing in
dynamic spectrum access networks,” IEEE Transactions on Computers, vol. 59, no. 7, pp.
865-877, 2010.
73