InfoMINDS: An Interdisciplinary Framework for Leveraging Data
Science upon Big Data in Surface Mining Industry
Vitor Afonso Pinto
1 a
and Fernando Silva Parreiras
2 b
1
Technology Department, Operational Technology for Base Metals South Atlantic, Vale, Carajas, Para, Brazil
2
Laboratory for Advanced Information Systems, FUMEC University, Rua do Cobre, Belo Horizonte, Brazil
Keywords:
Data Science, Big Data, Framework, Mining Industry.
Abstract:
Intending to be more and more data-driven, companies are leveraging data science upon big data initiatives.
However, to reach a better cost-benefit, it is important for companies to understand all aspects involved in such
initiatives. The main goal of this paper is to provide a framework that allows professionals from the mining
industry to accurately describe data science upon big data. The following research question was addressed:
”Which essential components characterize an interdisciplinary framework for data science upon big data in
mining industry?”. To answer this question, we will extend OntoDIVE ontology to create a framework capable
of explaining aspects involved in such initiatives for the mining industry. As a result, this paper will present
InfoMINDS - A Framework for Data Science upon Big Data Relating People, Processes and Technologies
on Mining Industry. This paper will contribute to leveraging data science initiatives upon big data allowing
application of OntoDIVE on real-case scenarios in mining industry.
1 INTRODUCTION
Data Science can be defined as an approach to ex-
tract worthy insights from low-value data. Big Data
can be defined as an integrated ecosystem of tech-
nologies performing formal roles with the purpose to
create technical conditions for the delivery of value-
added applications based on data. Data science upon
big data is seen by organizations as a tool to improve
operational efficiency though it has strategic poten-
tial, drive new revenue streams and gain competitive
advantages (Sivarajah et al., 2017).
There is a considerable literature addressing major
concepts related to data science and big data. Some
studies proposed ways to characterize data science
upon big data using concepts of volume, velocity,
variety, validity, veracity, variability, visibility, ver-
dict and value (Sharma, 2017; Corea, 2016; Addo-
Tenkorang and Helo, 2016). Other studies proposed
ways to group big data technologies (Bari et al., 2014;
Ahlemeyer-Stubbe and Coleman, 2014; Murthy et al.,
2014). Other studies proposed ways to explain data
science processes (Corea, 2016; Maimon and Rokach,
2010; Abbott, 2014; Takurta et al., 2017).
a
https://orcid.org/0000-0002-2731-0952
b
https://orcid.org/0000-0002-9832-1501
Additionally, a range of literature exists suggest-
ing an interdisciplinary approach as a success factor
for data science initiatives (Corea, 2016; Forte, 2015;
Cady, 2017; Luis, 2017). Other studies presented
results generated by data science initiatives that in-
volved multiple knowledge areas (Xu et al., 2014;
Fisher et al., 2017; Roy et al., 2017; Capalbo et al.,
2017; Zhou et al., 2016; Arias and Bae, 2016; San-
toro et al., 2018; Hurwitz et al., 2015; Van Der Aalst,
2016; Zheng et al., 2015; Tal
´
on-Ballestero et al.,
2018; Balliu et al., 2016; Lei et al., 2016; Maciejew-
ski, 2017; Lu and Li, 2017; Stoet and Geary, 2018;
Seele, 2017; Lubchenco and Grorud-Colvert, 2015).
Even with all the progress that has been made,
mining industry is still grappling with how to cap-
ture insights that are not obvious. As an example,
mining personnel not always understand how data sci-
ence initiatives are conducted in other industries. An-
other aspect is that mining companies not always have
expertise to define and deploy big data technological
ecosystems. Although big data comprises technolo-
gies performing formal roles, technologies may vary
among organizations and each technology should be
minutely chosen to avoid loss of effectiveness. Thus,
there is a risk of receiving biased advisory from con-
sultants who try to push technologies based on their
own interests or limitations.
784
Pinto, V. and Parreiras, F.
InfoMINDS: An Interdisciplinary Framework for Leveraging Data Science upon Big Data in Surface Mining Industry.
DOI: 10.5220/0010484107840791
In Proceedings of the 23rd International Conference on Enterprise Information Systems (ICEIS 2021) - Volume 2, pages 784-791
ISBN: 978-989-758-509-8
Copyright
c
2021 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
Considering the lack of concepts as a major con-
tributing factor for preventing the leverage of data sci-
ence initiatives upon big data, it is crucial to explain
all concepts related to this kind of initiative and the
interactions between and among them. The main goal
of this paper is to propose a conceptual framework
that allows professionals from the mining industry to
accurately describe data science upon big data.
Our approach differs from others as we intend to
create common ground so that data science initia-
tives upon big data can be fully understood by pro-
fessionals from any area of knowledge. In order
to contribute to the body of knowledge, this paper
is more interested in the general idea or conception
behind data science initiatives upon big data rather
than any individual instance of those initiatives. In
this context, the following research question was ad-
dressed: ”Which essential components characterize
an interdisciplinary framework for data science upon
big data in mining industry?”.
By addressing this research question, in this paper
we propose InfoMINDS which is a conceptual frame-
work that organizes practices commonly applied dur-
ing data science initiatives upon big data, consider-
ing a comprehensive and end-to-end perspective. This
framework may contribute either to clarification of
concepts or the explanation of interactions between
and among them. InfoMINDS can also be considered
as a foundation upon which mining companies can
build policies, standards, rules, procedures, method-
ologies or any other artifact in order to leverage data
science initiatives upon big data. This paper is struc-
tured as follows: Section 1 presents the context of this
work. Section 2 presents methods of research. Sec-
tion 3 presents results that are discussed in Section 4.
Section 5 concludes the paper.
2 METHODS
A conceptual framework may be defined as an end
result of bringing together a number of related con-
cepts to explain or predict a given event or to give
a broader understanding of the phenomenon of in-
terest (Imenda, 2014). It is a visual presentation of
key variables, factors or concepts and their relation-
ship among each other which have been or have to
be studied (Miles and Huberman, 1994). The main
purpose of a conceptual framework is to bring focus
on the content and to act as a link between litera-
ture, methodology and results. We decided to build
a conceptual framework as it provides understanding,
rather than offering a theoretical explanation (Jaba-
reen, 2009).
In order to guarantee completeness and correct-
ness of InfoMINDS Framework, we decided to im-
plement it using OntoDIVE Ontology, presented in
(Pinto and Parreiras, 2020). In practical terms, we
created individuals on class Frameworks to repre-
sent either the InfoMINDS framework itself and each
one of its dimensions. We also created individuals on
class Processes to represent each of InfoMINDS pro-
cesses. As this study is focused on mining industry,
we created individuals on Class Frameworks to rep-
resent either the mining industry and each one of its
production phases (mining and mineral processing).
3 RESULTS
3.1 InfoMINDS Structure
InfoMINDS is structured in twenty-five processes that
are grouped in ve dimensions. Processes derived
from existing multidisciplinary literature. Dimen-
sions derived from ICT business processes: Plan,
Build, Run, Enable and Manage. InfoMINDS is de-
signed to be flexible as each initiative is unique. Fig-
ure 1 presents InfoMINDS Framework. Next subsec-
tions present details on dimensions and processes. We
made InfoMINDS available at GitHub
1
to encourage
its usage in future studies.
3.1.1 Dimension A: Plan
Dimension A (Plan). includes processes that aim at
understanding and defining the goals of end users and
the environment in which data science initiative will
take place. Table 1 presents the five processes in-
cluded in this dimension along with their major goals.
3.1.2 Dimension B: Build
Dimension B (Build). includes processes to select-
ing, preprocessing, transforming data and also model-
ing and evaluating data applications. Table 2 presents
the five processes included in this dimension along
with their major goals.
3.1.3 Dimension C: Run
Dimension C (Run). includes processes to deploy,
manage and monitor data application and the out-
comes generated by them. Table 3 presents the three
processes included in this dimension along with their
major goals.
1
https://github.com/tecladista1/InfoMINDS
InfoMINDS: An Interdisciplinary Framework for Leveraging Data Science upon Big Data in Surface Mining Industry
785
Table 1: Processes from Dimension A (Plan).
Process Major Goals
Acquire
Interdisciplinary
Team
To provide an interdisciplinary
team capable of executing all
required activities to achieve
results proposed by data sci-
ence initiatives. This process
focus on explanation of some
functions that need to be per-
formed so that a data science
initiative can be successful.
Understand
Business
Context
This process involves under-
standing the goals of the end-
user in terms of what is ex-
pected from data science ini-
tiative, considering all existing
constraints and requirements.
Business Process Modeling is
an approach that could be used
to facilitate this process.
Define
Appropriate
Paradigms
To choose appropriate
paradigms for conducting
a data science initiative con-
sidering the current scenario.
Depending on the expected re-
sults a different paradigm may
be chosen: agile, waterfall,
among others.
Define
Technological
Ecosystem
To define a technological
ecosystem to support a data
science initiative. Although
Big Data comprises technolo-
gies performing formal roles,
the technologies chosen to
perform each role may vary
among organizations or even
among initiatives.
Determine
Initiative
Readiness
To determine the maturity of
people, processes and tech-
nologies to conduct a data sci-
ence initiative upon big data.
This process intends to clar-
ify if all conditions are set to
starting a data science initia-
tive upon big data.
*Source: Authors.
3.1.4 Dimension D: Enable
Dimension D (Enable). includes processes to ad-
dress computing infrastructure, procurements, cy-
bersecurity and other enabling processes. Table 4
presents processes included in this dimension and
their goals.
Table 2: Processes from Dimension B (Build).
Process Major Goals
Perform Data
Selection
This process consists in iden-
tifying data sources, acquir-
ing, integrating and transfer-
ring data. Data need to be con-
sistently aggregated from dif-
ferent sources of information,
and integrated with other sys-
tems and platforms.
Perform Data
Preprocessing
Much of the raw data con-
tained in databases is un-
preprocessed, incomplete, and
noisy. The main goal of this
process is to treat outliers, in-
consistent values, missing val-
ues, redundant fields and obso-
lete fields.
Perform Data
Transformation
The main goal of this pro-
cess is to transform or consol-
idate data so that the resulting
data science processes may be
more efficient. Data transfor-
mation comprehends transfor-
mation, dimension reduction
and discretization of data.
Perform Data
Modeling
The main goal of this pro-
cess is to create a model
based on initial hypothesis, ex-
ploratory data analysis, clas-
sification, clustering, among
others. Models can be equa-
tions linking quantities that we
can observe or measure. They
can also be a set of rules.
Perform Model
Evaluation
The main goal of this process
is to perform validation and
verification tests of data appli-
cation. Model evaluation is the
process of assessing a prop-
erty or properties of a model
in terms of its structure and
data inputs so as to determine
whether or not the results can
be used in decision-making
*Source: Authors.
3.1.5 Dimension E: Manage
Dimension E (Manage)., includes processes to ad-
dress strategical processes, portfolio management,
risks, among others. Table 5 presents processes in-
cluded in this dimension and their goals.
ICEIS 2021 - 23rd International Conference on Enterprise Information Systems
786
Table 3: Processes from Dimension C (Run).
Process Major Goals
Deploy Data
Application
The main goal of this process
is to deploy authorized version
of data application in a produc-
tion environment.
Manage Data
Application
The main goal of this process
is to manage data application,
including incidents, problems,
changes, among others.
Monitor Data
Application
Outcomes
The main goal of this process
is to capture outcomes pro-
vided by data application.
*Source: Authors.
Table 4: Processes from Dimension D (Enable).
Process Major Goals
Develop and
Manage Team
To manage and develop team
allocated to a particular data
science initiative.
Deploy and
Improve
Computing
Infrastructure
To deploy technologies for
data creation, acquisition,
transmission, ingestion, stor-
age, pre-processing, data
modeling, among others.
Manage and
Control
Procurements
To provide contracts with ex-
ternal vendors and partners.
Implement and
Monitor Data
Governance
To implement and monitor the
maturity level of data gover-
nance practices.
Implement and
Monitor
Information
Security
To implement and monitor
policies and routines to pre-
vent or mitigate risks related to
information security.
Manage and
Control Budget
To manage and control eco-
nomic and financial budgets.
*Source: Authors.
3.2 InfoMINDS Application
In this section we present how we applied Info-
MINDS Framework to data science initiatives upon
big data on the mining industry. We used InfoMINDS
to organize activities of four data science initiatives:
two initiatives focused on operations and two fo-
cused on maintenance. Initiatives followed the se-
quence indicated by InfoMINDS Dimensions: Plan,
Build, Run, Enable and Manage. To ensure correct-
ness and completeness of this stage, we used On-
toDIVE ontology, presented in (Pinto and Parreiras,
2020) to represent all elements involved on initiatives.
Table 5: Processes from Dimension E (Manage).
Process Major Goals
Provide
Strategy
Alignment
To share business strategy
with all data science initiatives
upon big data.
Manage
Overall
Portfolio
To manage overall portfolio of
either projects or services..
Manage and
Control Risks
To identify, assess, manage
and control risks.
Manage and
Control
Resources
To manage and control non-
financial resources available
for all initiatives.
Manage
Benefits
Realization
To manage benefits realization
for all the initiatives.
Share
Knowledge and
Information
To share knowledge and infor-
mation about previous initia-
tives.
*Source: Authors.
3.2.1 Implementation of Dimension A: Plan
Following guidelines of Processes A.1 Acquire In-
terdisciplinary Team, A.2 Understand Business
Context, A.3 Define Appropriate Paradigm, A.4
Define Technological Ecosystem and A.5 Deter-
mine Initiative Readiness, interdisciplinary teams
were allocated to each initiative. Next, each squad
was assigned to a real business problem and teams
chose the most appropriate paradigm for the initiative
under their responsibility. Next, teams searched On-
toDIVE Ontology to find existing technologies pre-
viously implemented by mining processes specialists.
At the end, teams decided to move all initiatives to the
next stage after considering their readiness.
3.2.2 Implementation of Dimension B: Build
Following guidelines of Processes B.1 Perform Data
Selection, B.2 Perform Data Preprocessing, B.3
Perform Data Transformation, B.4 Perform Data
Modeling and B.5 Perform Model Evaluation, all
teams performed activities to select, preprocess and
transform data from available data sources. Next,
data models were created and evaluated. Actions per-
formed in this stage were different as each initiative
had a different starting point. While some initiatives
were focused on rolling out existing applications to
a different location, other initiatives had to build pre-
dictive models and applications from the scratch.
InfoMINDS: An Interdisciplinary Framework for Leveraging Data Science upon Big Data in Surface Mining Industry
787
3.2.3 Implementation of Dimension C: Run
Following guidelines of Processes C.1 Deploy Data
Application, C.2 Manage Data Application, C.3
Monitor Data Application Outcomes, all teams per-
formed activities to deploy and manage data appli-
cations besides monitoring outcomes of these data
applications. We created individuals on Class Out-
comes to represent data applications deployed in this
stage. These data applications were linked to roles in
the big data technological ecosystem through Object
Property haveRole.
3.2.4 Implementation of Dimension D: Enable
Following guidelines of Processes D.1 Develop and
Manage Team, D.2 Deploy and Improve Comput-
ing Infrastructure, D.3 Manage and Control Pro-
curements, D.4 Implement and Monitor Data Gov-
ernance, D.5 Implement and Monitor Informa-
tion Security and D.6 Manage and Control Budget,
team “TE00 Squad Shared Services” oversaw peo-
ple development, management of computing infras-
tructure, management of budget and procurements
and monitoring of data governance and information
security. Initiatives started exchanging files and as
the time went on, improvements were made to enable
real-time acquisition of data.
3.2.5 Implementation of Dimension E: Manage
Following guidelines of Processes E.1 Provide Strat-
egy Alignment, E.2 Manage Overall Portfolio, E.3
Manage and Control Risks, E.4 Manage and Con-
trol Resources, E.5 Manage Benefits Realization
and E.6 Share Knowledge and Information, this di-
mension comprehends processes to either start or fin-
ish initiatives. To start the initiatives analyzed by this
study, first they were authorized. Then, all financial
and non-financial resources were made available. We
used object property supportedBy to link initiatives
to people, processes and technologies.
3.3 InfoMINDS Examples
This section presents outcomes generated by initia-
tives analyzed in this paper. Examples of this sec-
tion were included to illustrate the potential of Info-
MINDS Framework.
3.3.1 Data Science for Mining Operations
This initiative was created to improve energy effi-
ciency in mining operations, by reducing fuel con-
sumption. In practical terms, this initiative intended
to address the following business problem: In what
extension the fuel consumption on coal mining opera-
tions is affected by other variables?. Table 6 presents
outcomes of this initiative in order to illustrate Info-
MINDS capabilities.
3.3.2 Data Science for Mineral Processing
Mineral processing includes size reduction and en-
richment of minerals. This initiative was created
to improve coal marketability, by ensuring values of
yield above 30% and ash below 11.2%. In practical
terms, this initiative intended to address the following
business problem: In what extension ash and yield
are affected by other variables of coal mineral pro-
cessing?. Table 7 presents outcomes of this initiative
in order to illustrate InfoMINDS capabilities.
3.3.3 Data Science for Mining Maintenance
Industrial maintenance intends to guarantee availabil-
ity and reliability for facilities and equipment. This
initiative was created to improve asset management,
by extending lifetime of mining trucks. In practical
terms, this initiative intended to address the follow-
ing question: In what extension the lifetime of mining
trucks are affected by other process variables?. Ta-
ble 8 presents outcomes of this initiative in order to
illustrate InfoMINDS capabilities.
3.3.4 Data Science for Plant Maintenance
Plant maintenance seeks for optimum availability,
optimum operating conditions, maximum utiliza-
tion of maintenance resources, optimum equipment
life, minimum spares inventory and ability to react
quickly. This initiative was created to improve asset
management, by reducing tearing on conveyor belts.
In practical terms, this initiative intended to address
the following question: In what extension tearing of
conveyor belts are related to other process variables?.
Table 9 presents outcomes of this initiative in order to
illustrate InfoMINDS capabilities.
4 DISCUSSION
InfoMINDS is a conceptual framework that orga-
nizes practices commonly applied to design, build
and maintain data applications considering a compre-
hensive and end-to-end perspective. InfoMINDS cre-
ates a common vocabulary allowing processes for de-
velopment and maintenance to be reused and shared
amongst industries from different segments.
ICEIS 2021 - 23rd International Conference on Enterprise Information Systems
788
E - Manage
D - Enable
Develop and Manage Team
D.1
Implement and Monitor
Data Governance
D.4
Deploy and Improve
Computing Infrastructure
D.2
Implement and Monitor
Information Security
D.5
Manage and Control Procurements
D.3
Manage and Control Budget
D.6
Provide Strategy Alignment
E.1
Manage and Control Resources
E.4
Manage Overall Portfolio
E.2
Manage Benefits Realization
E.5
Manage and Control Risks
E.3
Share Knowledge and Information
E.6
A - Plan
Acquire Interdisciplinary Team
A.1
Define Technological Ecosystem
A.4
Understand Business Context
A.2
Define Appropriate Paradigms
A.3
B - Build
Perform Data Selection
B.1
Perform Data Modeling
B.4
Perform Data Preprocessing
B.2
Perform Data Transformation
B.3
C - Run
Deploy Data Application
C.1
Manage Data Application
C.2
Monitor Data Application Outcomes
C.3
Perform Model Evaluation
B.5
Determine Initiative Readiness
A.5
Figure 1: InfoMINDS: Framework Overview.
*Source: Authors.
Table 6: Examples of Data Science for Mining Operations.
Predictive
Application
A predictive application was built to identify processes variables that could influence fuel
consumption based on routes, conditions of roads and inclination angles of equipments.
Prescriptive
Application
A prescriptive application was built and deployed into the fleet management system which
is used by operators to receive information from dispatch controllers.
Descriptive
Application
A descriptive application was built and deployed to facilitate finding deviations to recom-
mendations of predictive application.
*Source: Authors.
InfoMINDS may contribute either to clarification of
concepts or to the explanation of interactions between
and among them. It can also be considered as a foun-
dation upon which mining companies can build poli-
cies, standards, rules, procedures, methodologies or
any other artifact in order to leverage data science ini-
tiatives upon big data.
InfoMINDS allows professionals from any knowl-
edge area to understand major processes and activi-
ties related to data science initiatives upon big data
as well as the processes and activities related to the
management, maintenance and support of data appli-
cations. All initiatives analyzed in this study were led
by personnel with no previous knowledge about big
data technologies or data science management.
InfoMINDS contributes either to clarification of
concepts or to the explanation of interactions between
and among them. Mining industry can benefit from
InfoMINDS as transaction technologies, previously
implemented by mining processes specialists, gen-
erate large volumes of scattered data and integrated
analyses of those data may be used as a tool for im-
proving operational efficiency of the industry. Info-
MINDS takes all existing technologies into consid-
eration as they have the potential to be used as data
source for data science initiatives. Besides that, Info-
MINDS helps mining industry to share best practices
between different sites. Data applications showed in
this paper could be rolled out to similar operations,
leveraging data science upon big data.
5 CONCLUSION
InfoMINDS framework is a conceptual framework
designed to create common ground so that data sci-
ence initiatives upon big data can be fully understood
by professionals from any area of knowledge. It has
twenty-five processes grouped into five dimensions
and contributes to leveraging data science initiatives
InfoMINDS: An Interdisciplinary Framework for Leveraging Data Science upon Big Data in Surface Mining Industry
789
Table 7: Examples of Data Science for Mineral Processing.
Predictive
Application
A predictive application was built and deployed to identify processes variables that could
influence results of ash and yield. This application was designed to identify the physical
type of coal being processed and automatically suggest parameters so that results of ash and
yield could be inside an expected range.
Prescriptive
Application
A prescriptive application was built and deployed into the system used by operators to de-
termine setups for processes variables. This prescriptive application brought tangible results
as operators were able to see the distance between their setups and the optimal ranges.
Descriptive
Application
A descriptive application was built and deployed to facilitate finding deviations to recom-
mendations of predictive application. This application was used to monitor adherence by
operator, by shift, and so on.
*Source: Authors.
Table 8: Examples of Data Science for Mining Maintenance.
Predictive
Application
A predictive application was built and deployed to identify variables that could influence
lifetime of mining trucks. This application analyzes fueling data, haul truck telemetry,
engineering parameters and laboratory results to recommend the sequence of scheduled
maintenance.
Prescriptive
Application
A prescriptive application was built and deployed into the maintenance workshop. This
application is focused on presenting trucks with lifetime lower than expected and suggests
the components to be replaced to extend the lifetime of equipment.
Descriptive
Application
A descriptive application was built and deployed in order to facilitate finding deviations to
recommendations of predictive application. This application is used to monitor components
suggested to be replaced.
*Source: Authors.
Table 9: Examples of Data Science for Plant Maintenance.
Predictive
Application
A predictive application was built and deployed to identify variables that could quickly
detect tearing of conveyor belts. After analyzing different process variables, a single process
variable was identified as capable of indicating the start of a tearing event.
Prescriptive
Application
A prescriptive application was built and deployed into the plant controller to automatically
stop conveyor belts whenever requested by predictive application.
*Source: Authors.
upon big data in mining industry. It can be considered
as a foundation upon which mining companies can
build policies, standards, rules, procedures, method-
ologies or any other artifact, in order to leverage data
science. It also may help mining industry profession-
als to draw parallels between data science results for
a different domain to their own domain.
InfoMINDS confirmed its capability of explaining
interactions between people, processes and technolo-
gies in the context of data science upon big data on
mining industry. The framework confirmed its con-
tribution to the clarification of concepts and termi-
nologies related to either data science or big data in
mining industry. In this study, all initiatives were led
by personnel with no previous knowledge about data
science. Still, consistent data science results were
achieved. InfoMINDS showed it can enlarge possi-
bilities of data science applications.
This study has several limitations. Firstly, In-
foMINDS is based on OntoDIVE ontology and in-
herits its limitations. Besides that, InfoMINDS was
conceived as a conceptual framework. Assuming
that different researchers may approach a single phe-
nomenon using different perspectives, it is possi-
ble that they might end up with different conceptual
frameworks as final result. Another limitation is the
fact InfoMINDS was applied in data science initia-
tives of a single mining company. Future works could
apply InfoMINDS on more real-case scenarios to col-
lect insights and thoughts of more people. Future
works could also implement a system based on OWL
file generated by Proteg
´
e.
ICEIS 2021 - 23rd International Conference on Enterprise Information Systems
790
REFERENCES
Abbott, D. (2014). Applied predictive analytics: principles
and techniques for the professional data analyst. Wi-
ley, Indianapolis (IN).
Addo-Tenkorang, R. and Helo, P. T. (2016). Big data ap-
plications in operations/supply-chain management: A
literature review. Computers & Industrial Engineer-
ing, 101:528 – 543.
Ahlemeyer-Stubbe, A. and Coleman, S. (2014). A practical
guide to data mining for business and industry. John
Wiley & Sons.
Arias, M. B. and Bae, S. (2016). Electric vehicle charging
demand forecasting model based on big data technolo-
gies. Applied energy, 183:327–339.
Balliu, A., Olivetti, D., Babaoglu,
¨
O., Marzolla, M., and
S
ˆ
ırbu, A. (2016). A big data analyzer for large trace
logs. Computing, 98(12):1225–1249.
Bari, A., Chaouchi, M., and Jung, T. (2014). Predictive
Analytics for Dummies. Wiley, [s.l].
Cady, F. (2017). The data science handbook. John Wiley &
Sons, Inc., Hoboken, NJ.
Capalbo, S. M., Antle, J. M., and Seavert, C. (2017). Next
generation data systems and knowledge products to
support agricultural producers and science-based pol-
icy decision making. Agricultural systems, 155:191–
199.
Corea, F. (2016). Big data analytics: a management per-
spective, volume 21. Springer, [s.l].
Fisher, J. B., Melton, F., Middleton, E., Hain, C., Ander-
son, M., Allen, R., McCabe, M. F., Hook, S., Baldoc-
chi, D., Townsend, P. A., et al. (2017). The future of
evapotranspiration: Global requirements for ecosys-
tem functioning, carbon and climate feedbacks, agri-
cultural management, and water resources. Water Re-
sources Research, 53(4):2618–2626.
Forte, R. M. (2015). Mastering predictive analytics with R:
master the craft of predictive modeling by developing
strategy, intuition, and a solid foundation in essential
concepts. Packt Publishing, Birmingham, UK.
Hurwitz, J., Kaufman, M., and Bowles, A. (2015). Cogni-
tive Computing and Big Data Analytics. Wiley Pub-
lishing, [s.l], 1st edition.
Imenda, S. (2014). Is there a conceptual difference between
theoretical and conceptual frameworks? Journal of
Social Sciences, 38(2):185–195.
Jabareen, Y. (2009). Building a conceptual framework:
philosophy, definitions, and procedure. International
journal of qualitative methods, 8(4):49–62.
Lei, Y., Jia, F., Lin, J., Xing, S., and Ding, S. X. (2016). An
intelligent fault diagnosis method using unsupervised
feature learning towards mechanical big data. IEEE
Transactions on Industrial Electronics, 63(5):3137–
3147.
Lu, H. and Li, Y. (2017). Artificial intelligence and com-
puter vision. Springer.
Lubchenco, J. and Grorud-Colvert, K. (2015). Making
waves: The science and politics of ocean protection.
Science, 350(6259):382–383.
Luis, T. (2017). Data mining with R: learning with case
studies. CRC Press, Taylor and Francis Group, Boca
Raton.
Maciejewski, M. (2017). To do more, better, faster and
more cheaply: Using big data in public administra-
tion. International Review of Administrative Sciences,
83(1 suppl):120–135.
Maimon, O. and Rokach, L. (2010). Data mining and
knowledge discovery handbook. Springer, New York.
Miles, M. B. and Huberman, A. M. A. M. (1994). Quali-
tative data analysis: an expanded sourcebook. Sage
Publications, Thousand Oaks, Calif., 2nd ed. edition.
Murthy, P., Bharadwaj, A., Subrahmanyam, P. A., Roy, A.,
and Rajan, S. (2014). Big data taxonomy.
Pinto, V. A. and Parreiras, F. S. (2020). Ontodive: An ontol-
ogy for representing data science initiatives upon big
data technologies. In ICEIS (1), pages 42–51.
Roy, S., Ray, R., Roy, A., Sinha, S., Mukherjee, G., Pyne,
S., Mitra, S., Basu, S., and Hazra, S. (2017). Iot, big
data science & analytics, cloud computing and mo-
bile app based hybrid system for smart agriculture.
In 2017 8th Annual Industrial Automation and Elec-
tromechanical Engineering Conference (IEMECON),
pages 303–304. IEEE.
Santoro, G., Fiano, F., Bertoldi, B., and Ciampi, F. (2018).
Big data for business management in the retail indus-
try. Management Decision.
Seele, P. (2017). Predictive sustainability control: A review
assessing the potential to transfer big data driven ‘pre-
dictive policing’ to corporate sustainability manage-
ment. Journal of Cleaner Production, 153(C):673–
686.
Sharma, K. (2017). Overview of Industrial Process Automa-
tion. Elsevier Science.
Sivarajah, U., Kamal, M. M., Irani, Z., and Weerakkody, V.
(2017). Critical analysis of big data challenges and
analytical methods. Journal of Business Research,
70:263 – 286.
Stoet, G. and Geary, D. C. (2018). The gender-
equality paradox in science, technology, engineering,
and mathematics education. Psychological science,
29(4):581–593.
Takurta, D., Ericson, G., Martens, J., and Severtson, B.
(2017). Deployment stage of the team data science
process lifecycle - azure.
Tal
´
on-Ballestero, P., Gonz
´
alez-Serrano, L., Soguero-Ruiz,
C., Mu
˜
noz-Romero, S., and Rojo-
´
Alvarez, J. L.
(2018). Using big data from customer relationship
management information systems to determine the
client profile in the hotel sector. Tourism Management,
68:187–197.
Van Der Aalst, W. (2016). Data science in action. In Process
Mining, pages 3–23. Springer.
Xu, L. D., He, W., and Li, S. (2014). Internet of things in
industries: A survey. IEEE Transactions on Industrial
Informatics, 10(4):2233–2243.
Zheng, X., Chen, W., Wang, P., Shen, D., Chen, S., Wang,
X., Zhang, Q., and Yang, L. (2015). Big data for so-
cial transportation. IEEE Transactions on Intelligent
Transportation Systems, 17(3):620–630.
Zhou, K., Fu, C., and Yang, S. (2016). Big data driven smart
energy management: From big data to big insights.
Renewable and Sustainable Energy Reviews, 56:215–
225.
InfoMINDS: An Interdisciplinary Framework for Leveraging Data Science upon Big Data in Surface Mining Industry
791