Explainable Business Intelligence for Video Analytics in Retail

Christian Daase

, Christian Haertel

and Klaus Turowski

Institute of Technical and Business Information Systems, Otto-von-Guericke University, Magdeburg, Germany

Keywords: Artificial Intelligence, Business Intelligence, Video Analytics, Society 5.0, Data Science.

Abstract: This paper explores research questions and perspectives for the next stage of societal development, often

referred to as Society 5.0, and the field of modern retail. Artificial intelligence (AI) is seen as a key component

that provides retailers with the means to optimize their store layouts, advertising campaigns, and overall

business strategy. The need to make AI-based decisions comprehensible and tangible to ensure acceptance by

the respective target groups has been emphasized with the concept of explainable AI in various research

works. Based on observations from the AI domain and the business world, the need to integrate explainability

into commercial AI-driven operations is addressed and the concept of explainable business intelligence (XBI)

is proposed. A set of potential research questions for video analytics in retail in the age of Society 5.0, as one

of the most promising use cases in this regard, is derived from the literature and the proposal of XBI in terms

of the outlined opportunities and challenges is explained, critically discussed, and visualized.

1 INTRODUCTION

Currently, the larger scheme of artificial intelligence

(AI) is permeating society at large as it is increasing

its influence on fundamental areas of science,

economics, governments, and social life. While it is

not uncommon for technological developments to be

accompanied by shifts in social structures and norms,

morality, or laws (Jiang et al. 2022), AI is expected to

have an impact that goes beyond minor adjustments

of communities to the shaping of a new social

revolution referred to as Society 5.0 (Carayannis and

Morawska-Jancelewicz 2022; Muslikhin et al. 2021;

Nair et al. 2021). The idea, also often mentioned as

smart society, leads back to a publication from the

Japanese government and describes a futuristic

human-centered society in which the virtual and

physical spaces are converging by means of AI, big

data analytics, robotics, virtual and augmented

reality, the internet of things (IoT), and other

technologies that are connected to them (Carayannis

and Morawska-Jancelewicz 2022; Muslikhin et al.

2021). However, although technological advances of

this magnitude are generally considered positive,

communities tend to express concerns, including

https://orcid.org/0000-0003-4662-7055

https://orcid.org/0009-0001-4904-5643

https://orcid.org/0000-0002-4388-8914

fears of potential job losses, social isolation, criminal

acts by entities unaware of moral and ethical

principles, and seamless surveillance (Elliott et al.

2021).

In recent years, studies provided indicators that

AI will most likely have the largest value impact

across the technological landscape on the retailing

sector (Guha et al. 2021). In turn, the retailing

industry is expected to benefit the most from

advancements in the field of AI (Bellis and Johar

2020). Since demographics change due to a new

generation of digitally educated humans, thus

facilitating the evolution towards new societal

concepts, retailers need to adapt their business

practices (Kahn et al. 2018). One especially

promising subfield of AI with manifold use cases in

retailing is video analytics, which can take place

either in-store (Kahn et al. 2018; Kaur et al. 2020;

Liciotti et al. 2017) or detached from physical

locations, for example by analyzing prerecorded

video content (Agrawal and Mittal 2022; Zhang et al.

2020). The analysis of critical business data in

conjunction with the required technological systems

and practices with the goal of understanding the

business and ultimately making beneficial and timely

decisions is also referred to as business intelligence

784

Daase, C., Haertel, C. and Turowski, K.

Explainable Business Intelligence for Video Analytics in Retail.

DOI: 10.5220/0012694600003690

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 26th International Conference on Enterprise Information Systems (ICEIS 2024) - Volume 1, pages 784-791

ISBN: 978-989-758-692-7; ISSN: 2184-4992

(BI) (Chen et al. 2012). As a connector between data

and its targeted use, the term data science has

emerged as an umbrella term for quantitative and

qualitative methods developed to make predictions

and solve data-related problems (Waller and Fawcett

2013). To draw a clear line in the context of this

paper, AI is considered the overarching term for

technologies that artificially replicate human

intelligence. Thus, it is the umbrella scheme behind

video analytics, which in turn is expected to be an

integral part of the next evolutionary stage of business

intelligence when applied in a commercial context

such as retail. With the projected convergence of AI-

related technologies and humans in Society 5.0, video

analytics systems could become more widespread,

while acceptance by stakeholders must be ensured.

When it comes to utilizing data from private

persons, even if recorded in public areas, privacy

concerns are a closely connected issue. In order to

mitigate reluctance in the adoption of AI-driven

technologies, the concept of explainable AI has

emerged to make artificially determined decisions

comprehensible for the average user (Bailao

Goncalves et al. 2022; Jiang et al. 2022). In this paper,

the definitions and characteristics of explainable AI

as necessary part of Society 5.0 and the grander

scheme of business intelligence are attempted to be

merged under the term explainable business

intelligence (XBI). By carefully approaching the

future of business data analysis, reluctance is

intended to be avoided from the outset. The focus of

this preliminary investigation of XBI is to position

video analytics in retail as a potential use case and to

describe critical questions for future studies. In

summary, this research aims to provide an answer for

the following research question (RQ):

RQ: What are potential RQs that can be derived

from the AI-based use case of video analytics

in retail under consideration of societal shifts

towards Society 5.0?

2 LITERATURE REVIEW

To create a knowledge base for the establishment of

the potential RQs in Section 4 and the concept of XBI

in Section 5, the conduction of a systematic literature

review (SLR) is described in this section. The SLR

was conducted using four peer-reviewed journals

from the database ScienceDirect of the publisher

Elsevier that focus on retail, commerce, marketing,

and generally technological solutions for the business

sector. The selected journals are Electronic

Commerce Research and Applications, International

Journal of Research in Marketing, Journal of

Retailing, and Journal of Retailing and Consumer

Services. As a secondary source of literature, the

abstract and citation database Scopus was consulted,

which claims to be the largest database of its kind

(Kitchenham and Charters 2007). This addition is

intended to ensure that potentially relevant literature

from other reliable scientific sources is not omitted if

it is not covered in any of the primary journals

reviewed. Additionally, SpringerLink with its own

selection of high-quality research articles (that might

not be indexed in Scopus) is considered as another

literature source.

To limit the number of articles, certain inclusion

and exclusion criteria were defined. First, the time

frame to be considered was set to the publication

period between January 2017 and October 2023 to

ensure actuality. Test searches led to the conclusion

that a combination of terms related to retail and video

yield the most appropriate results for the individual

databases. The exact queries are stated in Table 1.

Table 1: Literature sources and specifications.

Source Restrictions

All - Period: Jan 2017 - Oct 2023

ScienceDirect - marke

as research article

Electronic Commerce

Research and

pplications

---

International Journal

of Research in

arketing

- Abstract contains retail*

ournal o

etailin

---

Journal of Retailing

and Consume

Services

- Title contains retail or

retailing

SpringerLink - Title contains retail,

retailing, or retailer

- Text contains video

rticle o

con

erence

Scopus - Title contains retail*

- Abstract contains video

rticle o

conference pape

Further selection criteria were set for the two review

phases of reading the abstracts and reading the full

texts in detail. In the first phase, duplicates were

removed, including semantic duplicates from the

same authors within a short period of time, as well as

articles consisting only of an introduction or missing

the abstract. Moreover, articles limited to a too

specific geographical region were rejected as it was

assumed that they potentially lack generalizability.

Finally, articles without an application area that is

related to video analytics were removed.

Explainable Business Intelligence for Video Analytics in Retail

785

In the second phase, articles whose full texts were

unavailable were removed from the list along with

publications that did not provide an outlook on future

data-related technology impacts or that did not

provide a suitable research justification in general.

Articles lacking a technical business perspective were

rejected as well. Lastly, from the remaining articles,

a dedicated selection of the most relevant material

was used in terms of this present paper. Figure 1

shows the review process, including criteria and

numbers of remaining articles after each stage.

Figure 1: Literature review workflow visualization.

3 BACKGROUND - DATA

SCIENCE IN BUSINESS

The societal shifts through the ongoing digitalization

have increased the generated amount of data

significantly (Yin and Kaynak 2015). As this trend is

expected to continue in the future, enterprises with

access to a high volume of data seek ways to make

use of the data to achieve performance improvements

(Chen et al. 2012; Wamba et al. 2017). Accordingly,

it can be stated that data is becoming the “most

valuable asset for any organization” (Nielsen 2017).

In turn, the discipline data science (DS) has gained

additional attention over the last years. DS is defined

as „methodology for the synthesis of useful

knowledge directly from data through a process of

discovery or of hypothesis formulation and

hypothesis testing“ (Chang and Grady 2019). Since

DS is mainly applied in the context of datasets that

correspond to the big data paradigm, the term big data

science emerged. Generally speaking, DS can be

interpreted as all activities in an analytics pipeline to

derive insights from data (Chang and Grady 2019)

and it is regarded as a super-set of other related

disciplines such as (big) data analytics, data mining,

and statistics. Additionally, in the enterprise context,

DS is referred to as data science for business or

business intelligence, as stated in the introduction

(Medeiros et al. 2020; Newman et al. 2016).

However, there are other interpretations that view DS

as a subset of BI (Larson and Chang 2016). Foley and

Guillemette (2010) define BI as “a combination of

processes, policies, culture, and technologies for

gathering, manipulating, storing, and analyzing data

collected from internal and external sources, in order

to communicate information, create knowledge, and

inform decision making”, which contradicts the

observation of Larson and Chang (2016), as clear

conceptual differences become visible.

The benefits of DS for businesses can be

manifold. For example, Medeiros et al. (2020)

mention increased efficiency and effectiveness in

decision-making through better quality of data and

information (data quality), actionable insights by

analysis of resources and data visualization

(analytical intelligence), improved organizational

knowledge and detection of business opportunities

(dynamic capabilities), as well as enhanced

productivity, performance, and profitability

(competitive advantages). In order to realize these

advantages and provide managers with the relevant

recommendation and insights (Chen et al. 2012;

Waller and Fawcett 2013), the respective BI/DS

projects require successful completion. However,

these types of undertakings are prone to failure

(VentureBeat 2019) due to multi-faceted challenges

related to project management and technical aspects

(Martinez et al. 2021; Saltz and Krasteva 2022). As

DS contains big data and big data analytics (Newman

et al. 2016), using the “appropriate technology for the

different activities such as data storage and

processing as well as building analytical models”

(Haertel et al. 2023) and deploying the results is

required (Saltz and Shamshurin 2016).

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

786

4 VIDEO ANALYTICS IN RETAIL

Based on observations from the literature analysis,

RQs arising in the sector of video analytics in retail

can be divided into three categories: purpose-focused,

technical, and juridical. The reason for dividing the

research into these three dimensions is that by finding

appropriate answers to the overarching questions of

“What should be achieved?”, “How can it be

achieved?” and “What is allowed to be done without

potential harm?”, businesses can implement

successful DS-driven solutions. The following

explanations serve to answer the RQ posed at the

beginning and form the first contribution of this work,

before the subsequent XBI concept.

4.1 Purpose-Focused Questions

Video analytics in retail has already been

implemented in various areas, for example for

understanding customer behavior in-store (Kaur et al.

2020; Liciotti et al. 2017), marketing reasons (Xiao et

al. 2023; Zhang et al. 2020), or extracting sentiments

from review videos (Agrawal and Mittal 2022). To

enable a comprehensive view on the topic, the areas

that it can affect need to be placed in context. In

addition, suitable metrics need to be developed to

measure the influence of analytical approaches. Since

user acceptance is a fundamental prerequisite for the

adoption of AI technologies (Jiang et al. 2022),

another question that arises is which influence the

awareness of being analyzed has on the customers, as

it is conceivable that humans adjust their behavior in

that case. Summarizing the purpose-focused

questions, the interdisciplinarity of the transformation

of retail business models through AI, big data, and

associated technologies (Alexandrova and Kochieva

2021) can serve as a foundation for the question

which research disciplines apart from economic

research might benefit from advances in video

analytics in retail as well, such as social science.

4.2 Technical Questions

The technical perspective on this topic covers how to

store and process data independent of the specific

purpose or applicable law. However, legal regulations

and deployed technology can sometimes not clearly

be separated. Especially privacy issues and the

necessity to anonymize people and data force AI

developers to consider layers of data abstraction,

specifically for video material (Jabłonowska et al.

2018; Kopalle et al. 2022). Thus, one technical

question is how to encode data for anonymized

processing. Another challenge occurs specifically for

in-store analysis, strongly connected to the intended

purpose, in terms of camera positioning. For example,

approaches to track the general movements of

customers or shopping carts in a store (Ferracuti et al.

2019) require a different positioning than cameras for

eye tracking to assess which item the customer is

looking at and how long the visual inspection lasts

(Huddleston et al. 2018). Also, the circumstances and

timing of video analytics offer important RQs to be

answered. Difficulties from analyzing in-store video

data immediately (Kahn et al. 2018; Liciotti et al.

2017) or during live streaming e-commerce sessions

(Chen et al. 2023) differ from analyzing material

asynchronously, for example from customer review

videos (Agrawal and Mittal 2022). Before being able

to analyze video content, another question is related

to ways to efficiently gather and store such data.

Nowadays, developers are turning towards cloud

computing as a possible centralized solution to

storage issues (Liciotti et al. 2017) while sensors,

mobile devices and the IoT in general are utilized for

data collection, also to complement data captured

with cameras (Kahn et al. 2018; Muslikhin et al.

2021). Due to its range of cost-effective, flexible and

easily usable applications, cloud computing is

considered a key component of future retail in general

(Daase et al. 2023). Lastly on the list of hereby

identified potential RQs, the user acceptance needs to

be ensured. Hence, in special consideration of neural

networks in AI, another question for future research

is if and how optimization techniques for such

networks can be adapted to make neural networks

comprehensible throughout every iteration of

optimization for the developer and the targeted user

group.

4.3 Juridical Questions

The last category on the list for future RQs, juridical,

covers laws and regulations for video analytics in

general. The first part of the questions is two-fold,

depending on the mode in which videos are analyzed.

If data is gathered in an in-store scenario, one of the

most important aspects is ensuring the anonymity of

customers (Jabłonowska et al. 2018; Kopalle et al.

2022). If prerecorded video content is analyzed

(Agrawal and Mittal 2022), another question is which

regulations in terms of copyright are applicable. In

addition to the technical question how and where to

store data efficiently, the physical location of stored

data must be assessed regarding local laws for the

juridical perspective. Table 2 summarizes these

potential RQs for video analytics in retail scenarios,

Explainable Business Intelligence for Video Analytics in Retail

787

especially focusing on AI with neural networks, thus

proposing a preliminary answer to the initially posed

RQ of this research-in-progress paper.

Table 2: Potential RQs for video analytics in retail.

Category Potential research questions

Purpose-

focused

⋅ Which areas in retailing can be affected

by video analytics?

⋅ Which metrics could be used to measure

benefits in specific retailing areas?

⋅ How do customers adapt their behavior in

case of in-store video analytics if they are

aware of being analyzed?

⋅ How can other research disciplines apart

from economic research benefit from

video anal

tics in retail?

Technical

⋅ How should video data be encoded for

efficient anonymized processing?

⋅ How should cameras be places for

specific purposes for in-store analytics?

⋅ What technical difficulties arise for in-

store video analytics in contrast to

asynchronous analysis and vice versa?

⋅ Which technologies can be used for

gathering and storing video datasets?

⋅ How can optimization techniques for

neural networks be adapted to ensure user

acce

tance in im

rovement iterations?

Juridical

⋅ Which regulations have to be considered

for in-store video analytics (privacy)?

⋅ Which regulations apply when analyzing

publicly available content (copyright)?

⋅ Where can data and analytics results be

store

to compl

with local laws?

5 EXPLAINABLE BUSINESS

INTELLIGENCE

The stated questions are focused on one potential

research field of retail in the era of Society 5.0. The

accumulated definition, opportunities, and challenges

of the interconnection between AI, BI, and the

necessity to underpin both with explainability in the

near future support to the proposal of the concept of

explainable business intelligence (XBI). On the one

hand, BI encompasses techniques and processes to

gather and analyze business data from internal and

external sources to make beneficial and informed

decisions in a business context (Chen et al. 2012;

Foley and Guillemette 2010). On the other hand,

explainable AI is the aspiration to construct AI

applications that allow to understand how the

underlying AI works, which factors it takes into

account, and how decisions are derived, with the

ultimate goal of increasing user acceptance (Bailao

Goncalves et al. 2022; Jiang et al. 2022). However, it

is important to emphasize that technical

understanding is not the only way to realize XAI, as

AI solutions and developments may exceed the

average human knowledge in this field. Instead, the

solutions and decisions made by AI could be

validated for correctness and suitability. In this way,

the need to understand the functioning of components

of AI such as neural networks could be alleviated.

The definition of XBI developed here is that XBI

represents the philosophy that all business data

retrieved during the production, delivery and retail

process must either be analyzed by means that can be

understood by human intelligence, or at least the

results must be verifiable using tools by humans with

limited technical knowledge. Furthermore, the

analysis results must be presented in such a way that

the average target group and beneficiaries can make

informed decisions about whether to trust the results.

As stated, an essential component of the proposed

concept of XBI is the possibility of checking the

results for their trustworthiness.

Developers of business applications in Society

5.0 can adapt their techniques to comply with the

argued concept of XBI. For example, AI-based

software could be developed with greater

involvement of humans whose expertise is to be

emulated, as they can be seen as design templates for

these systems (Daase and Turowski 2023). A second

approach might be to test AI applications more

intensively and to make the test procedures public,

including inputs and (expected) outputs. Thus, for the

more technically experienced audience, the

development process and focused intentions of the

organization providing the system can become clearer.

A third constituent of XBI can be to involve the target

audience in the result evaluation. The AI system can

be tested with artificial data that is based on

suggestions by the user group, making sure that the

system’s capabilities match the expectations. Finally,

as a fourth component, results and AI-driven

decisions can be verified by simulations and

mathematical analyses, as a supplement or even as a

substitute if the technical traceability of the inner

workings of AI solutions cannot be realized.

Figure 2 illustrates an abstracted data flow

towards XBI. First, data is collected as usual, for

example from various sources in different formats,

with varying velocities, and in varying volumes, thus

complying with the concept of big data. In connection

with the analysis of this data in a business context by

any means yields business intelligence. If the data is

analyzed by an explainable artificial intelligence

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

788

system, incorporating the aforementioned

components, the concept of explainable business

intelligence can be realized.

Figure 2: Abstracted data flow towards XBI.

6 DISCUSSION

The evolution from traditional BI to XBI can be

considered a natural process as well as a necessity in

the modern digitalized business world. While in the

past, mathematical models, manual data collection,

dedicated analytics units and extensive discussions

with decision makers were common for gaining

business knowledge, AI solutions are increasingly

taking over (parts of) the mentioned tasks (Jiang et al.

2022). Thus, more and more vital actions for

organizations are influenced by systems to which

humans have limited access to and which, in

comparison to human intelligence and decision

making, can partly not be explained.

The contribution of this paper is discussed on the

basis of a SWOT analysis (strengths, weaknesses,

opportunities, threats). First, the outcome of this

study, meaning the concept of XBI and the questions

that necessarily need to be answered when applying

the concept to retail video analytics, is considered

from the perspective of its potential positive impact

when deployed and considered internally (strengths).

It can be argued that using the XBI concept to analyze

business data with AI technologies in a way that

enables human decision makers to understand either

the entire analysis process or at least the verification

of the results could more easily convince stakeholders

to invest in and trust AI to a greater extent. This in

turn would lead to more effective and widespread

adoption and thus an increase in retail revenue, as

retail is the industry that is expected to offer some of

the greatest potential for AI and is one of the biggest

beneficiaries of AI, as mentioned in the introduction

(Bellis and Johar 2020; Guha et al. 2021). However,

as the concept is generic, this strength can also apply

to various other business areas. As for the potential

RQs for video analytics listed in Table 2, most of the

questions on evaluation metrics, behavioral science,

configuration, and legally compliant implementation

could be answered much more easily when posed in

a setting suitable for XBI.

Weaknesses of the XBI concept become

conceivable when considering the current state of AI

and the enthusiasm for it. With a wealth of data now

readily available as a basis for machine learning

models and computing power reaching extraordinary

levels (Eling et al. 2022; Jiang et al. 2022), many

companies are trying to quickly exploit the potential

in this highly dynamic field. If they now switch from

mature AI systems to a different approach with a far

more human-centric focus, companies could be

forced to rethink key parts of their digitalization

strategy. In doing so, they risk increased costs and

falling behind competitors if the applied XBI concept

does not lead to increased business effectiveness,

customer loyalty, or other monetarily rewarding

compensations.

The opportunities of the XBI concept depend on

the occurrence of external events and unpredictable

factors such as human behavior. Regarding the

specific use case of video analytics in retail, some

believe that people are becoming unaware of the

constant surveillance by cameras and therefore

implicitly accept it (Elliott et al. 2021), which would

prevent unusual customer behavior due to perceived

surveillance. However, it is assumed that trust in AI

itself is still insufficient (Jiang et al. 2022). This

would mean that the problem is not distrust in the

acquisition of video material, as it could just be a store

detective reviewing the footage, but the exploitation

of inherent knowledge by systems that cannot be

looked into. XBI has the potential to bridge this gap

by convincing customers to engage in AI-driven

business by making them understand how the system

Explainable Business Intelligence for Video Analytics in Retail

789

works, how the results are verified, and what

consequences they lead to.

Finally, threats are also caused by external factors

that impose undesirable circumstances on an

organization. As indicated, such a threat could be a

reaction from the target group (i.e., stakeholders or

customers) that has a negative impact on the potential

benefits of XBI. One fear of unexplained AI and one

fear of XAI can be considered as examples

respectively. On the one hand, the fear that an AI

system can make decisions that are in some way

harmful without adequate monitoring is frequently

expressed. On the other hand, individuals from the

human target group (e.g., customers recorded on

video) may fear that the involvement of humans in the

AI system will compromise their privacy more than if

a mindless machine analyzes the recorded data in a

black box. If the second fear outweighs the first,

people may become more reluctant towards XAI

approaches, not because they do not trust the human-

centered development process, but because they do

not trust the humans themselves who are involved in

it. As a result, the implementation of XBI could be

controversial.

This SWOT analysis underlines that the concept

of XBI still offers plenty of room for discussion.

Particularly with regard to video analytics in retail,

XBI could be a sensible approach, even if it is still at

a preliminary conceptual stage.

7 CONCLUSION

In this research-in-progress article, a first literature

analysis was conducted to position the topic of video

analytics in retail as one expectable key component

of AI utilization in commercial business. A selection

of some of the most pressing issues for

implementation was elaborated from the literature to

identify potentials and challenges in this regard. With

the conceptualization of explainable business

intelligence as a link between explainable artificial

intelligence and traditional business intelligence, an

outlook on a promising new business digitalization

strategy with a human-centric focus was given and by

means of a SWOT analysis critically discussed.

In further research, the topics of Society 5.0 and

its connection to AI technologies are to be explored

in greater depth to find answers to the questions

developed in Table 2 in the retail sector. In addition,

the concept of XBI is to be shaped more precisely, as

reducing the reluctance in the adoption of AI

technologies is expected to be vital for future

businesses. The overall goal in finalizing this research

endeavor is to position XBI as a key component of the

future economy.

REFERENCES

Agrawal, S. R., & Mittal, D. (2022). Optimizing customer

engagement content strategy in retail and E-tail:

Available on online product review videos. In Journal

of Retailing and Consumer Services, 67, 102966.

Alexandrova, E., & Kochieva, A. (2021). Modern Aspects

of Digital Technologies Development in Retail

Networks. In Lecture Notes in Networks and Systems.

Comprehensible Science. Springer International

Publishing.

Bailao Goncalves, M., Anastasiadou, M., & Santos, V.

(2022). AI and public contests: a model to improve the

evaluation and selection of public contest candidates in

the Police Force. In Transforming Government: People,

Process and Policy, 16(4), 627–648.

Bellis, E. de, & Johar, G. V. (2020). Autonomous Shopping

Systems: Identifying and Overcoming Barriers to

Consumer Adoption. In Journal of Retailing, 96(1), 74–

87.

Carayannis, E. G., & Morawska-Jancelewicz, J. (2022).

The Futures of Europe: Society 5.0 and Industry 5.0 as

Driving Forces of Future Universities. In Journal of the

Knowledge Economy, 13(4), 3445–3471.

Chang, W. L., & Grady, N. (2019). NIST Big Data

Interoperability Framework: Volume 1, Definitions.

Chen, Chiang, & Storey (2012). Business Intelligence and

Analytics: From Big Data to Big Impact. In MIS

Quarterly, 36(4), 1165.

Chen, Q., Yan, X., Zhao, Y., & Bian, Y. (2023). Live

streaming channel strategy of an online retailer in a

supply chain. In Electronic Commerce Research and

Applications, 62, 101321.

Daase, C., & Turowski, K. (2023). Conducting Design

Science Research in Society 5.0 – Proposal of an

Explainable Artificial Intelligence Research

Methodology. In Lecture Notes in Computer Science.

Design Science Research for a New Society: Society 5.0.

Springer Nature Switzerland.

Daase, C., Volk, M., Staegemann, D., & Turowski, K.

(2023). The Future of Commerce: Linking Modern

Retailing Characteristics with Cloud Computing

Capabilities. In Proceedings of the 25th International

Conference on Enterprise Information Systems.

SCITEPRESS - Science and Technology Publications.

Eling, M., Nuessle, D., & Staubli, J. (2022). The impact of

artificial intelligence along the insurance value chain

and on the insurability of risks. In The Geneva Papers

on Risk and Insurance - Issues and Practice, 47(2),

205–241.

Elliott, K., Price, R., Shaw, P., Spiliotopoulos, T., Ng, M.,

Coopamootoo, K., & van Moorsel, A. (2021). Towards

an Equitable Digital Society: Artificial Intelligence

(AI) and Corporate Digital Responsibility (CDR). In

Society, 58(3), 179–188.

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

790

Ferracuti, N., Norscini, C., Frontoni, E [E.], Gabellini, P.,

Paolanti, M., & Placidi, V. (2019). A business

application of RTLS technology in Intelligent Retail

Environment: Defining the shopper's preferred path and

its segmentation. In Journal of Retailing and Consumer

Services, 47, 184–194.

Foley, É., & Guillemette, M. G. (2010). What is Business

Intelligence? In International Journal of Business

Intelligence Research, 1(4), 1–28.

Guha, A., Grewal, D., Kopalle, P. K., Haenlein, M.,

Schneider, M. J., Jung, H., Moustafa, R., Hegde, D. R.,

& Hawkins, G. (2021). How artificial intelligence will

affect the future of retailing. In Journal of Retailing,

97(1), 28–41.

Haertel, C., Pohl, M., Nahhas, A., Staegemann, D., &

Turowski, K. (2023). A Survey of Technology

Selection Approaches for Data Science Projects. In

AMCIS 2023.

Huddleston, P. T., Behe, B. K., Driesener, C., & Minahan,

S. (2018). Inside-outside: Using eye-tracking to

investigate search-choice processes in the retail

environment. In Journal of Retailing and Consumer

Services, 43, 85–93.

Jabłonowska, A., Kuziemski, M., Nowak, A. M., Micklitz,

H.‑W., Pałka, P., & Sartor, G. (2018). Consumer Law

and Artificial Intelligence: Challenges to the EU

Consumer Law and Policy Stemming from the

Business' Use of Artificial Intelligence - Final report of

the ARTSY project. In SSRN Electronic Journal.

Jiang, Y., Li, X [Xiang], Luo, H., Yin, S., & Kaynak, O.

(2022). Quo vadis artificial intelligence? In Discover

Artificial Intelligence, 2(1), Article 4.

Kahn, B. E., Inman, J. J., & Verhoef, P. C. (2018).

Introduction to Special Issue: Consumer Response to

the Evolving Retailing Landscape. In Journal of the

Association for Consumer Research, 3(3), 255–259.

Kaur, J., Arora, V., & Bali, S. (2020). Influence of

technological advances and change in marketing

strategies using analytics in retail industry. In

International Journal of System Assurance Engineering

and Management, 11(5), 953–961.

Kitchenham, B., & Charters, S. (2007). Guidelines for

performing Systematic Literature Reviews in Software

Engineering: Technical report, Version 2.3 EBSE

Technical Report.

Kopalle, P. K., Gangwar, M., Kaplan, A., Ramachandran,

D., Reinartz, W., & Rindfleisch, A. (2022). Examining

artificial intelligence (AI) technologies in marketing via

a global lens: Current trends and future research

opportunities. In International Journal of Research in

Marketing, 39(2), 522–540.

Larson, D., & Chang, V. (2016). A Review and Future

Direction of Agile, Business Intelligence, Analytics and

Data Science. In International Journal of Information

Management 36(5).

Liciotti, D., Frontoni, E [Emanuele], Mancini, A., &

Zingaretti, P. (2017). Pervasive System for Consumer

Behaviour Analysis in Retail Environments. In Lecture

Notes in Computer Science. Video Analytics. Face and

Facial Expression Recognition and Audience

Measurement. Springer International Publishing.

Martinez, I., Viles, E., & Olaizola, I. G. (2021). Data

Science Methodologies: Current Challenges and Future

Approaches. In Big Data Research 24.

Medeiros, M. M. de, Hoppen, N., & Maçada, A. C. G.

(2020). Data science for business: benefits, challenges

and opportunities. In The Bottom Line, 33(2), 149–163.

Muslikhin, M., Horng, J.‑R., Yang, S.‑Y., Wang, M.‑S., &

Awaluddin, B.‑A. (2021). An Artificial Intelligence of

Things-Based Picking Algorithm for Online Shop in the

Society 5.0's Context. In Sensors (Basel, Switzerland),

21(8).

Nair, M. M., Tyagi, A. K., & Sreenath, N. (2021). The

Future with Industry 4.0 at the Core of Society 5.0:

Open Issues, Future Opportunities and Challenges. In

2021 International Conference on Computer

Communication and Informatics (ICCCI). IEEE.

Newman, R., Chang, V., Walters, R. J., & Wills, G. B.

(2016). Model and experimental development for

Business Data Science. In International Journal of

Information Management, 36(4), 607–617.

Nielsen, O. B. (2017). A Comprehensive Review of Data

Governance Literature. In Selected Papers of the

IRIS(Issue Nr 8).

Saltz, J., & Krasteva, I. (2022). Current approaches for

executing big data science projects - a systematic

literature review. In PeerJ Computer Science, 8(e862).

Saltz, J., & Shamshurin, I. (2016). Big data team process

methodologies: A literature review and the

identification of key factors for a project's success. In

2016 IEEE International Conference on Big Data (Big

Data). IEEE.

VentureBeat. (2019). Why do 87% of data science projects

never make it into production? https://venturebeat.com/

2019/07/19/why-do-87-of-data-science-projects-never-

make-it-into-production/

Waller, M. A., & Fawcett, S. E. (2013). Data Science,

Predictive Analytics, and Big Data: A Revolution That

Will Transform Supply Chain Design and Management.

In Journal of Business Logistics, 34(2), 77–84.

Wamba, S. F., Gunasekaran, A., Akter, S., Ren, S. J., Dubey,

R., & Childe, S. J. (2017). Big data analytics and firm

performance: Effects of dynamic capabilities. In

Journal of Business Research, 70, 356–365.

Xiao, L., Li, X [Xiaofeng], & Zhang, Y. (2023). Exploring

the factors influencing consumer engagement behavior

regarding short-form video advertising: A big data

perspective. In Journal of Retailing and Consumer

Services, 70

, 103170.

Yin, S., & Kaynak, O. (2015). Big data for modern industry:

challenges and trends [point of view]. In Proceedings

of the IEEE, 103(2), 143–146.

Zhang, H., Li, Y., Ai, Q., Luo, Y., Wen, Y., Jin, Y., & Ta,

N. B. D. (2020). Hysia: Serving DNN-Based Video-to-

Retail Applications in Cloud. In Proceedings of the

28th ACM International Conference on Multimedia.

ACM.

Explainable Business Intelligence for Video Analytics in Retail

791