Semantic Micro-Front-End Approach to Enterprise Knowledge

Graph Applications Development

Milorad Tosic

1,2 a

, Nenad Petrovic

and Olivera Tosic

Faculty of Electronic Engineering, University of Nis, Aleksandra Medvedeva 14, Nis, Serbia

Virtuona d.o.o, Nis, Serbia

Keywords: IoT, Low-Code, Ontologies, Semantic Micro-Front-Ends, Mapping, Enterprise Knowledge Graph.

Abstract: Industry 4.0 has been mainly driven by IoT devices and artificial intelligence developments rising

heterogeneity of the data acquired by sensing devices as well as data from existing legacy systems (such as

ERP) the crucial for digital transformation. Until recently, migration of enterprise applications to Cloud has

been considered the only viable long-term solution. However, after hidden infrastructure costs of the Cloud-

only based approach have been discovered, a number of businesses have begun considering hybrid Cloud-

Edge architectures where Micro-Services Architectures (MSA) on backend are complemented with Micro-

Font-End (MFE) applications. However, the architecture must be very carefully optimized in order to avoid

high risks and costs due to increased system’s complexity. In this paper, a semantic-driven approach based

on Enterprise Knowledge Graph (EKG) and ontologies with their automated mapping is introduced in order

to manage the complexity. Ontologies are adopted for automated, low-code approach to composition and

deployment of MFE components targeting enterprise productivity applications. MFE applications generated

this way are built upon Semantic Micro Services backend that can transparently be distributed between Cloud

and Edge. Our approach is illustrated on the case study for semantic annotation of manufacturing area which

utilizes a shared marketplace component for IoT-based indoor positioning.

1 INTRODUCTION

During the past decade, Internet of Things (IoT) and

artificial intelligence (AI) have emerged becoming

key enablers of novel usage scenarios and innovation

across various application domains. Industry 4.0 is

certainly one of them aiming to accelerate and make

more efficient manufacturing as well as other

industry-related activities (Babu et al., 2023). While

IoT assumes acquisition of huge data amounts thanks

to incorporation of large number of smart wearable

and sensing devices, artificial intelligence strives to

analyse the acquired data in order to extract useful

knowledge. The extracted knowledge can be further

leveraged for reasoning and decisions which aim to

improve not only the manufacturing process but high-

level enterprise effectiveness as well. However,

development of such applications and services faces

several challenges (Petrovic et al., 2019). First, there

are many different types of devices involved, each of

https://orcid.org/0000-0001-8142-5788

https://orcid.org/0000-0003-2264-7369

them possibly using different data formats. Then, in

Industry 4.0 applications, it is traditionally needed to

integrate these data streams with the data originating

from existing industrial platforms, including legacy

systems and ERPs (Alsaadi, 2022). Finally,

deployment of such services often constraints some

parts to be strictly executed on Edge servers located

within the enterprise premises (such as

collecting/sensing potentially sensitive data), while,

the acquired data is often offloaded to Cloud for

heavy processing tasks, such as artificial intelligence

empowered predictive analytics (Petrovic et al.,

2019). In order to overcome such bottlenecks,

artificial intelligence techniques have been identified

as one of the crucial methods (Romero et al., 2023).

This paper aims to bridge the Cloud-Edge gap

with the goal to make development of state-of-the-art

industry-oriented applications more effective. The

proposed approach makes use of ontologies and

semantic technology in synergy with Micro-Font-End

488

Tosic, M., Petrovic, N. and Tosic, O.

Semantic Micro-Front-End Approach to Enterprise Knowledge Graph Applications Development.

DOI: 10.5220/0012236200003584

In Proceedings of the 19th International Conference on Web Information Systems and Technologies (WEBIST 2023), pages 488-495

ISBN: 978-989-758-672-9; ISSN: 2184-3252

(MFE) components in order to cope with the target

systems’ complexity. Semantic technology is adopted

for integration of various data sources together with

domain conceptualization, application component

composition, configuration and customization. In the

same time, MFE components are leveraged to make

deployment of the no-code platform more flexible

and capable of adopting Cloud as well as Edge

services. The novelty of this combination is the high

degree of automation of software development and

deployment steps, making it usable even by

professionals without IT background.

The proposed approach is implemented in the

ASMOSA (Agile Semantic Model driven

development for Smart Applications ecosystem in

manufacturing) software platform, which adopts

semantic approach to information systems design.

ASMOSA enables users to create a no-code/low-code

applications tailored to the specific needs of

individual enterprise operations in various business

domains. In our case study, we make use of WiPos,

one of existing SHOP4CF components (Zimniewicz,

2020) aiming indoor localization for implementation

of manufacturing area monitoring application.

2 BACKGROUND AND RELATED

WORKS

2.1 Ontologies and Semantic

Technology

As domain conceptualization, ontologies have been

widely used in practice for knowledge representation

across many domains – biology, healthcare, public

administration, Internet of Things, computing

infrastructure management to business and enterprise

information systems. In these cases, ontologies make

various usage scenarios possible, including

interoperability of heterogeneous devices, data

integration, reasoning against large amount of

semantically represented information to discover

hidden knowledge and code generation based on

specific parameters (Petrovic, Tosic, 2020).

In this paper, ontologies are leveraged to tackle

various aspects that make development of MFE

applications development more convenient (semantic

repository annotation, application model), while we

also make use of them in order to create unified

semantic knowledge based using data from various

sources (ERP, IoT devices). Furthermore, our work

makes use of ontology alignment, also known as

ontology matching. In this work, a set of proprietary

ontologies (modelling production, planning, and

transportation) are used for alignment with domain-

specific ontologies. This way, knowledge sharing and

reuse is facilitated.

2.2 MSA and MFE Applications

Development

Micro services-based architecture (MSA) represents

paradigm where application is developed as a

collection of services, which can be developed,

deployed and maintained independently (Pontarolli et

al., 2023). MFEs build upon the underlying concepts

of MSA, while their main idea is to think about a web

application as a composition of features owned by

independent teams (Taibi et al., 2022, Geers, 2023).

Apps created using MFE components should be

independent, self-contained and do not share state or

global variables.

In this work, MFE-based approach is leveraged as

enabler for semantic-driven composition of micro

services. This way, execution of different

components is made possible either in Cloud or on

Edge, considering the isolation and independence of

runtimes for distinct MFEs. This way, deployment

flexibility is improved making possible industry-

oriented scenarios where sensor data is collected

within the Edge, but sent to Cloud only for

demanding processing tasks.

2.3 Shared Component Marketplaces

The approach proposed in this paper builds upon the

concept of shared component marketplace as

proposed by several EU funded projects to address

complexity of enterprise applications development

(FIWARE 2023, MARKET4.0 2023, SHOP4CF

2023, Zimniewicz 2020). The case study application,

presented in the paper, SHOP4CF components

repository is adopted. IoT-based indoor localization

system using UWB is implemented by the WiPos

component (Wi-POS, 2023). The WiPos hardware

consists of a sink node connected via USB to

Raspberry Pi, 4 anchor nodes and a tag device which

position is calculated. WiPos pushes the calculated

relative position data from IoT devices to Orion

Context Broker instance. Orion Context Broker

(Orion Context Broker, 2023) is middleware aiming

convenient data acquisition from IoT devices relying

on publish-subscribe mechanism. Indoor positioning

is performed periodically and corresponding

information is published as NGSI v2 entity (Fonseca

et al., 2023) with the following properties: 1) id – tag

Semantic Micro-Front-End Approach to Enterprise Knowledge Graph Applications Development

489

identifier, 2) observedAt – timestamp, and 3)

relativePosition – [x, y] coordinates.

2.4 Related Works

There are several existing works that aim to enable

convenient low/no-code development of MSA-based

applications by tackling some of the relevant aspects,

relying on semantic technology or other underlying

approaches as well. Overview of these solutions

together with their descriptions and covered aspects

are given in Table 1.

Table 1: Overview of similar solutions.

olution

Description Aspect

SMADA-

Fog

(Petrovic et

al., 2019)

Model-driven, semantic-

enabled deployment of

containerized micro

services, supporting both

Edge and Cloud.

Deployment

Formaloo

(Formaloo,

2023)

Building customer portals,

CRMs, and other business

apps without any code.

App creation

and

deployment

AppSheet

(AppSheet,

2023)

Low-code development of

cross-platform mobile

apps, using tabular data

sources

App creation

and

deployment

of mobile

applications

Ontopic

Studio

(Ontopic,

2023)

Environment for

designing semantic layers

as a knowledge graph

with no code.

Domain

modelling

It can be seen that existing approaches often do

not include convenient integration of data from

various sources and usage of external service. On the

other side, we aim to cover all the mentioned aspect

within one solution. Our proposed approach aims to

tackle the mentioned challenges in area of Industry

4.0 enabling automated deployment of complex

industry-oriented applications, bridging the gaps

related to heterogeneity of devices and integration

with legacy systems. However, our approach also

aims to provide unified interoperability and

integration of external services based on their

specification, such as in case of OpenAPI.

Additionally, our ecosystem provides low-code

approach to incorporation of domain modelling and

conceptualization using Domain editor. Finally, apps

created using our solution based on MFE are mobile

device-compatible as well.

3 IMPLEMENTATION

Figure 1 depicts logical software architecture of the

proposed ASMOSA’s ecosystem.

Figure 1: System architecture overview.

The central component of ASMOSA platform is

Knowledge editor and runtime (KERT) application

which gives ability to domain experts (such as

operation managers in different departments -

manufacturing, sales, marketing, development, etc.)

to intuitively configure different individual

applications in a drug-n’-drop manner without any

programming-related knowledge. The key enabler for

easy-to-use, automated application creation using

KERT is semantic technology in synergy with

predefined MFE components. MFE components can

be imported from semantically annotated central

component repository, integrating various sources

(such as SVN or Git). When it comes to underlying

technology, the core of semantic engine is

implemented in Java relying on a triplestore

implementation of SPARQL endpoint, while MFEs

are implemented in HTML and JavaScript related

technologies.

On the other side, adoption of domain ontologies

provides convenient way to construct applications by

leveraging conceptualization, provided by domain

experts. Moreover, with respect to given semantic

representations, we provide mechanisms which

enable import of tabular data from various external

sources, such as ERP. Additionally, ontologies are

also used for semantic mapping of API specifications

(such as OpenAPI), which provides convenient

WEBIST 2023 - 19th International Conference on Web Information Systems and Technologies

490

Figure 2: Component interaction within the ASMOSA case study application.

extension with new services. This way, seamless

integration of data, services and knowledge available

on Edge servers as well as public or private Cloud

services is achieved.

When it comes to app creation using KERT, the

desired components can be imported from central

components repository and added to user’s project.

Furthermore, it is possible to customize UI elements,

such as colors of buttons, controls and labels. Once

the app design is finished, build and deployment can

be done, so the application becomes immediately

available to the end-users. Required back-end

functionality is provided by semantic integration of

different sources, including enterprise knowledge

graphs, simple data tables and interfaces of services.

This way, we also provide the means for leveraging

the existing SHOP4CF components and their

adoption within our MFEs. End-users simply access

available applications using either smartphone, tablet

or PC devices.

Additionally, we implemented BrokerConnector

Service which acts as general adapter between

TASOR-based applications and FIWARE Orion

Context Broker, relying on TASOR-compliant

libraries and service architecture. Its methods can be

used to interact with any instance of FIWARE Orion

Context Broker server specified by IP address as API

call parameter. BrokerConnector service methods are

application independent and can be used for data

collection in different use-case specific applications.

BrokerConnector service provides set of data

retrieval and management operations: creation of new

entities, creation of new subscriptions, handling of

entity updates utilizing publish-subscribe

mechanism, updating entities, deleting entities and

subscriptions. In this way, interaction of ASMOSA

platform and SHOP4CF infrastructure is achieved, as

illustrated in diagram within Figure 2 gives more

detailed view focused on interaction of the involved

components in the presented case study.

4 SEMANTIC FRAMEWORK

In this section, we give overview of the underlying

ontologies behind our approach, including both

domain-specific ontologies and auxiliary ontologies

which are further used for semantic mapping.

Figure 3: Overview of the proposed semantic-driven

approach to low code application development.

Illustration showing the role of ontologies within

our proposed approach is given in Figure 3, while the

part focused on MFEs is depicted in Figure 5.

ASMOSA is based on the TasorSCAS platform

(TasorSCAS, 2023) that represents all system

metadata semantically using ontologies. Its aim is to

adopt a common ontology to encode the shared

understanding. The corresponding communication

language is built upon this understanding and

implemented using the common ontology. Thanks to

that, each participant can “understand” the

communication language by alignment of the

common ontology of interaction to its own local

knowledge represented by one or more ontologies.

For that purpose, we make use of semantic mapping,

which is defined using a mapping language specified

by an ontology as well. In our case study, mapping is

used to cover the following aspects: 1) integration

with functionality of external services leveraging

OpenAPI specification – such as SHOP4CF

components 2) representation of domain-specific data

- such as JSON with relative position or ERP. In order

to unify the domain knowledge about various aspects

related to manufacturing, proprietary TasorSCAS

Manufacturing Ontologies (Production, Planning and

Transportation) are used. On the other side, we

provide Domain Editor where users can provide

additional information useful for integration of

external data using GUI.

Finally, once all the knowledge is grounded to

common base – semantic knowledge graph grounded

Semantic Micro-Front-End Approach to Enterprise Knowledge Graph Applications Development

491

Figure 4: Indoor localization Domain Ontology - an illustrative segment.

on Tasor Ontologies. Knowledge from this graph is

further used for generation of micro frontend

components that are finally composed into end-user

applications, where different parts of semantic

knowledge graph are leveraged for rendering various

parts of the result.

Figure 5: Semantic micro frontend based architecture.

4.1 WiPos Domain Ontology

The main purpose of this ontology is to formalize the

understanding of relative positioning data coming

from IoT-based WiPos indoor localization system

within the presented ASMOSA case study. Sink node

calculates tag position, while it is connected to

Raspberry Pi sending messages containing the 2D

coordinates representing tag device position in JSON

format to Orion Context Broker. On the other side,

UWB-based anchor nodes are also part of localization

system and they participate in position calculation.

Each of anchor nodes has both ordinal number and

IEEE address in order to be distinguished. In order to

semantically interpret the position-related content

within ASMOSA, a domain ontology was developed

using TasorSCAS tools, as illustrated in Figure 4.

4.2 TasorSCAS Manufacturing

Ontologies Framework

Conceptual foundation of the TasorSCAS framework

related to manufacturing builds upon of the following

three ontologies: Production Ontology, Planning

Ontology, and Transportation Planning Ontology.

These ontologies enable us to effectively understand

and work with specific data in different domains.

The production ontology primarily deals with

definitions of concepts related to production

resources, encompassing units, products, machines,

and two types of actions: durative (actions that occur

over a specific duration) and instantaneous (actions

that occur at a specific moment). By utilizing the

production ontology, our system gains a

comprehensive understanding of the production-

related entities and their associated actions.

The planning ontology, on the other hand, focuses

on data related to product orders, schedules, product

IDs, customer IDs, and prioritization. This ontology

assists in managing and organizing the planning

aspects of our system, allowing for efficient handling

of product orders, scheduling operations, and

prioritizing tasks based on predefined criteria.

Furthermore, we have developed a transportation

planning ontology that works with geolocations and

relative positioning, including locations on a factory

floor. It incorporates indoor positioning technologies

like WiPos. By integrating geolocation and

positioning data, our system can optimize

transportation planning within the factory

environment, enhancing efficiency and logistics.

4.3 Semantic Mapping Approach

ASMOSA is built upon the semantic-oriented

TasorSCAS infrastructure and makes use of semantic

mapping as defined there. Despite that more than few

respectful approaches to model mapping have been

proposed and some of them are established as

standards, together with more general foundational

ontologies (such as BFO, DOLCE, UFO) to the best

WEBIST 2023 - 19th International Conference on Web Information Systems and Technologies

492

Figure 6: Example semantic mapping of JSON file for data exchange with FIWARE Context Broker.

of our knowledge, none of them is fully ontology

based. Additionally, we provide an additional tool -

Domain Editor, which enables users to describe their

data using simple and intuitive GUI. In this way, user-

specific data are grounded on our Production

Ontology, so several applications related to

manufacturing (planning, logistics, scheduling, etc.)

can be adopted by users without any additional

complexities. Our aim is to eliminate traditionally

complex, costly and timely implementation projects

that are main barrier for modern agile digital

transformation practices, which represents value add.

An illustrative example showing how the mapping

ontology is applied within our case study for data

exchange with broker, shown in Figure 6 .

4.4 Domain-Specific Mapping

Ontology

As a proof of concept, we developed DMV mapping

ontology representing set of facts generated by

implementation of our Mapping Ontology on data

from specific Enterprise Resource Planning (ERP)

software used in our testing partner DMV Company.

This ontology is specifically tailored for the

requirements of the partner’s data. The DMV

mapping ontology serves as an intermediary,

facilitating the import of data from their ERP system

and aligning it with the corresponding ontologies

within our system, namely the production ontology,

planning ontology, and transportation planning

ontology. Importing data relies on automated steps,

eliminating the need for manual interventions in order

to save the time and effort needed. Therefore, such

customer-specific ontologies represent the main

target for future extensions for various cases.

4.5 OpenAPI Mapping Ontology

In order to implement and add new services within

our system, we decided to develop our services in

accordance with the OpenAPI scheme. We are using

Open API Schema version 3.0 and have developed

domain and mapping ontology for this version. The

developed ontology adopts and customizes the

previous work (Aikaterini et al., 2020) with a goal to

better fit the task in hand. Note that the Orion Broker

component, that is the central component in the

SHOP4CF infrastructure, publishes its API compliant

to the OpenAPI standard. Component Orion Broker

Connector is used to facilitate data integration

between ASMOSA and Orion Broker component. As

such, it publishes its API compliant to the OpenAPI

as well. In this way, any existing SHOP4CF

component can be included into solutions generated

using ASMOSA. As a proof-of-concept, we develop

applications that use WiPos component that publishes

data to the Orion Broker while ASMOSA collects

these data automatically using our connector

component (see Figure 2). When creating the

ontology, we opted for OpenAPI's JSON format. The

key element in the ontology itself is represented by

the Schema Object class, which represents the root

object of JSON itself. From this class we then derive

basic Open API objects such as: Info, Services, Paths,

and Components. Depiction of corresponding

semantic representation for OpenAPI Schema object

is given in Figure 7. For each of them, we define the

OpenAPI Object domain class using properties of the

same name, which represents a single object within

OpenAPI. Then, for each class that represents the

basic objects in the OpenAPI configuration, we added

properties that characterize the given class. An

interesting problem concerning the OpenAPI

ontology mapping relates to the Paths class. Namely,

in this case identifier is implicitly defined. We solved

the problem by introduction of the PathItem class that

represents the adminscopeOf of the Paths class and

represents the domain of the pathPrefix and

pathString properties as well as properties for the

methods used to work with the service itself.

Semantic Micro-Front-End Approach to Enterprise Knowledge Graph Applications Development

493

Figure 7: Ontology segment for semantic representation of the OpenAPI Schema object.

5 CASE STUDY

Goal of the presented case study application is to

enable efficient workforce and operation monitoring

in a manufacturing company. For this purpose, we

make use of WiPos indoor localization and semantic

annotations. In our experiment, the final application

was set up within the manufacturing area of DMV

company in Niš, Serbia.

Video of the live demo in action is available on

YouTube

. Semantic Annotation view gives the

ability of adding information about the machines

involved in industrial process, based on their indoor

position within the manufacturing area. Indoor

position is collected by means of WiPos tag that

publishes its position to the SHOP4CF Orion Context

broker component. On the other side, Figure 8 shows

the Operation Monitoring view, whose goal is to

provide insight to the managers about the machines,

operations performed, usage and work hours spent by

employees assigned to them.

Figure 8: Operation Monitoring view, rendered on Android

tablet device.

In order to create such applications, ASMOSA’s

app studio was used. Moreover, Domain Editor as

used in order to define crucial concepts – machines,

operations and location. After that, managers were

https://youtu.be/ajMcmGThpYw

able to simply import tabular data about operations

and machines from ERP.

6 EVALUATION

For purpose of evaluation for our approach, several

relevant aspects are considered based on the

presented case study: data sources integrated, time

needed for KERT-aided app creation and additional

artifacts needed for extension of the approach in order

to use it for another case study if required. Overview

of experiments and results is given in Table 2.

Table 2: Evaluation aspects and achieved outcomes based

on DMV case study.

Aspect Outcome Efforts needed

Data sources

WiPos positioning

system, DMV’s

ERP

WiPos ontology

DMV ontology

App creation

Semantic

Annotator and

Operation

Monitoring

Domain

specification

using KERT

Conecptualization

of domain

Domain

ontologies

Around 9 min

for ontology

creation

BrokerConnector

customization

WiPos broker

connector

Around 5 min

for API call

parameterization

As it can be seen, for custom tailoring of industry-

oriented applications, less than 15 minutes overall

were needed for additional steps that would make it

possible to apply the proposed solution to other case

studies, which eliminates the need for developing

components from scratch. Just several minutes are

enough for domain expert to define crucial concepts

of the underlying domain using Domain Editor,

making the import of external data from new sources

possible. This way, it is possible to significantly

WEBIST 2023 - 19th International Conference on Web Information Systems and Technologies

494

speed up the overall software development process,

while overcoming the barriers of both data source and

external service integration, thanks to utilization of

ontologies and semantic mapping. Finally, to achieve

customization of the presented solutions for new

application, programming expertise is not a must,

thanks to adoption of no-code approach enabled by

combination of ontologies and frontend components.

7 CONCLUSION AND FUTURE

WORKS

In this paper, we introduced an approach to

semantics-aided development of industry-oriented

applications. As a proof of concept, we demonstrated

case study evaluated within a realistic usage scenario.

According to the results, the proposed solution has

proven effective in several aspects of importance for

industry-oriented applications: integration of data

from heterogeneous sources (IoT devices, ERP) as

well as execution environments (IoT devices on the

Edge and Cloud services). MFE has been proven to

speed up the whole process of application

development and deployment, while making the

software applications development learning curve

much less step, bringing it closer to domain experts

without programming experience. Finally, the

presented case study could be easily adapted for other

IoT data sources given the corresponding domain

ontology and broker connector API calls.

In future, we plan to explore the potential of

trending ChatGPT conversational agent and Large

Language Models (LLM) in order to integrate it

within our semantics-enabled framework. Our goal is

to automatize some of the ontology-related tasks,

such as automated domain ontology construction

based on textual or tabular sources; complex

SPARQL query generation with minimal or without

user intervention; automated service code generation.

ACKNOWLEDGEMENTS

This research work has received funding from the

European Union’s Horizon 2020 research and

innovation program under grant agreement No.

873087 (SHOP4CF) as one of the winners in OC2.

REFERENCES

Alsaadi, N. (2022). Modeling and Analysis of Industry 4.0

Adoption Challenges in the Manufacturing Industry.

Processes. 10. 2150 (2022). 10.3390/pr10102150.

AppSheet, https://about.appsheet.com/home/, last

accessed: 17/07/2023.

Babu, B., Nallasivam, J. D., Aravinth, S. (2023). Overview

of Industry 4.0. 15th International Conference on

Science, Technology and Management, pp. 14-16

(2023)

FIWARE, https://www.fiware.org/marketplace/ last

accessed: 17/07/2023.

Fonseca, J.; Márquez, F.; Jacobs, T. NGSIv2 API

Speciﬁcation—v2.0n. Available online: https://

ﬁware.github.io/speciﬁcations/ngsiv2/stable/ (accessed

on 26. September 2023).

Formaloo, https://www.formaloo.com/en/, last accessed:

17/07/2023.

Geers, M., What are Micro Frontends?, https://micro-

frontends.org/, last accessed 16/07/2023.

Aikaterini, K., Mainas, N., Petrakis, E. GM. "Ontology for

openapi rest services descriptions." In 2020 IEEE 32nd

International Conference on Tools with Artificial

Intelligence (ICTAI), pp. 35-40. IEEE, 2020.

MARKET4.0, Marketplaces, https://market40.eu/project/

marketplaces/index.html last accessed: 17/07/2023.

Ontopic Studio, https://ontopic.ai/en/ontopic-studio/, last

accessed: 17/07/2023.

Orion Context Broker. https://fiware-orion.readthedocs.io

(accessed on 18 July 2023).

Petrovic, N., Tosic, M.: SMADA-Fog: Semantic model

driven approach to deployment and adaptivity in fog

computing, Simulation Modelling Practice and Theory

Volume 101, 102033, 1-25 (2020). https://doi.

org/10.1016/j.simpat.2019.102033

Pontarolli, R.P., Bigheti, J.A., de Sá, L.B.R., Godoy, E.P.L.

(2023). Microservice-Oriented Architecture for

Industry 4.0. Eng 2023, 4, 1179-1197.

https://doi.org/10.3390/eng4020069

Romero, J. R., Medina-Bulo, I., Chicano, F.: Optimising the

Software Development Process with Artificial

Intelligence, Springer Singapore. (2023)

SHOP4CF, https://www.shop4cf.eu/, last accessed

2023/07/12.

Taibi D. and Mezzalira, L. (2022). Micro-Frontends:

Principles, Implementations, and Pitfalls. SIGSOFT

Softw. Eng.

TasorSCAS, https://www.virtuonasoft.com/scas.html, last

accessed 2023/07/24.

Wi-POS Indoor Localization, https://shop4cf.github.io/

communication-docs/partners/IMEC/Wi-POS.html, la

st accessed 2023/07/12.

Zimniewicz, M. Deliverable 3.2 - SHOP4CF Architecture.

2020. https://elk.adalidda.com/2021/05/SHOP4CF-

WP3-D32-DEL-210119-v1.0.pdf (accessed on 18 July

2023).

Semantic Micro-Front-End Approach to Enterprise Knowledge Graph Applications Development

495