A Systematic Literature Review on Hybrid Deep Learning Smart

Recommendation Systems

Kadek Cahya Dewi

, Putu Indah Ciptayani

and Putu Adriani Prayustika

Department of Business, Politeknik Negeri Bali, Badung Bali, Indonesia

Department of Electrical Engineering, Politeknik Negeri Bali, Badung Bali, Indonesia

Keywords: Systematic Literature Review, Deep Learning, Hybrid Deep Learning, Smart Recommendation Systems,

Recommendation Systems.

Abstract: The purpose of the research was to conduct a Systematic Literature Review (SLR) to understand about the

hybrid deep learning model on smart recommendation system. SLR was held based on guidelines from

Kitchenham & Charters. There were four research question about datasets, methods, programming language

and also evaluation parameters that was used in hybrid deep learning smart recommendation system. The

research started with 4931 articles from digital libraries namely ACM Digital Library, IEEE Explore,

ScienceDirect and also SpringerLink. After third layer filtering, there found 50 articles became the data of the

SLR. It can be concluded that the analysis result namely: (1) textual dataset was the most used dataset, (2)

the combination of Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) was the

most widely used as hybrid deep learning method, (3) python is widely used in experiments, and (4) accuracy,

precision, recall, and F1-Score were the most often used as evaluation parameters in smart recommendation

systems.

1 INTRODUCTION

Personalized recommendation system is not only

providing items or products that meet the interests of

different users according to their interests, but also

recommend new items that meet their interests.

Product recommendation systems can influence

buying behaviour, consumer preferences, user

experiences, and also sales (Zhang & Bockstedt,

2020) (Wu & Ye, 2020) (Turkut, Tuncer, Savran, &

Yılmaz, 2020). Recommendation systems usually use

predicting ratings or develop a list of product

rankings for each user. In general, there are three

types of recommendation systems, namely Content-

based (CB) recommendation, Collaborative Filtering

(CF), and Hybrid models (Benabderrahmane,

Mellouli, & Lamolle, 2018) (Wang, Zhang, Xue, Lu,

& Na, 2019). Content-based and collaborative

filtering recommendation systems have evolved in

the last ten years. Hybrid model is a collaboration of

CB and CF models. The hybrid model in the

https://orcid.org/0000-0002-8922-6908

https://orcid.org/0000-0002-6923-3852

https://orcid.org/0000-0001-8280-9601

recommendation system was developed with the aim

of optimizing the recommendation results.

Some algorithms in artificial intelligence can be

applied and combined to produce an optimal

recommendation system model. The clustering

algorithm can be applied to content-based

recommendations. Clustering attempts to group

large-scale data points into several categories

according to their attributes. Various traditional

clustering algorithms have been implemented

including Gaussian Mixture Model, nearest

neighbours, K-means, mean-shift, graph community

detection, and DBSCAN (Wang, Zhang, Xue, Lu, &

Na, 2019). In intelligent systems, the Self Organizing

Maps (SOM) algorithm can be used in the clustering

process to provide product recommendations. Self-

Organizing Maps is an unsupervised neural network

algorithm created by Kohonen. SOM is widely used

in multidimensional data clustering (Dewi &

Harjoko, 2010). A product can be recommended if it

is in the same cluster.

Dewi, K., Ciptayani, P. and Prayustika, P.

A Systematic Literature Review on Hybrid Deep Learning Smart Recommendation Systems.

DOI: 10.5220/0010942000003260

In Proceedings of the 4th International Conference on Applied Science and Technology on Engineering Science (iCAST-ES 2021), pages 181-188

ISBN: 978-989-758-615-6; ISSN: 2975-8246

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

181

Collaborative Filtering (CF) is applied to the

recommendation system to find groupings of things

that usually occur together, such as pants and belts

together in a market-basket analysis. In addition to the

classical apriori algorithm, deep learning algorithms

can be applied in this association pattern. Deep

learning is a development of neural network learning.

Deep learning is a specialized field in machine

learning that focuses on the representation of data and

adds successive learning layers to improve the

representation of input data (Benabderrahmane,

Mellouli, & Lamolle, 2018) (Djellali & Adda, 2020).

Deep neural network architecture can be used to

predict or recommend various things as in research

(Benabderrahmane, Mellouli, & Lamolle, 2018) (Jha,

Prashar, Long, & Taniar, 2020) (Xu Y. , et al., 2019).

Based on (Wang, Zhang, Xue, Lu, & Na, 2019) (Dewi

& Harjoko, 2010). The use of deep learning can make

the recommendation system performance better.

Several previous studies have developed a hybrid

recommendation system model with various

methods. Research (Nilashi, Ibrahim, & Ithnin, 2014)

used Adaptive Neuro-Fuzzy Inference Systems

(ANFIS) and SOM methods. Then research about the

tourism industry (Nilashi, Bagherifard, Rahmani, &

Rafe, 2017) using a combination of SOM and

Expectation Maximization (EM) methods for

clustering and then predicting the recommendations

using the ANFIS and Support Vector Regression

(SVR) method. Research (Xu Y. , et al., 2019) used a

modified RNN in the Slanderous User Detection

Recommender System. Convolutional RNN was also

used in research (Adiyansjah, Gunawan, &

Suhartono, 2019) for the Music Recommender

System.

The basic idea of this research is to conduct a

systematic literature review to understand about the

hybrid deep learning model on a smart

recommendation system. There is no literature review

that discusses the types of datasets used in hybrid

deep learning smart recommendation systems, trend

hybrid deep learning methods that used to build smart

recommendation systems, trend frameworks used to

build hybrid deep learning smart recommendation

systems, and evaluation parameters used to measure

the success of a hybrid deep learning smart

recommendation system.

2 RESEARCH METHODOLOGY

The research was a systematic literature review

research. A systematic literature review could be

explained as a research method and process for

identifying and critically appraising relevant research

with purpose to identify all empirical evidence that

fits the pre-specified inclusion criteria to answer a

particular research question or hypothesis (Snyder,

2019).

The research setting was digital libraries, namely

ACM Digital Libraries (from 2020-2021), IEEE

Explore (from 2020-2021), ScienceDirect (from the

last 5 year), and SpringerLink (from the last 5 years).

The data collection technique used is the

documentation of articles and journals. The research

instrument was a check-list for the classification of

research materials and research notes.

The data analysis technique was a Systematic

Literature Review (SLR) according to the guidelines

from Kitchenham & Charters (Kitchenham &

Charters, 2007). The analysis divided into three big

phase, which are planning, conducting and reporting

the review.

Research question was built with purpose to

maintain the focus of the literature review. This

condition facilitates the process of finding data

needed. The research question namely:

1. What kind of datasets are the most used for a

hybrid learning smart recommendation system?

(RQ1)

2. What kind of methods are the most used for a

hybrid learning smart recommendation system?

(RQ2)

3. What kind of programming languages are

proposed for a hybrid learning smart

recommendation system? (RQ3)

4. What kind of parameters are used for evaluating a

hybrid learning smart recommendation system?

(RQ4)

This study used several components, namely: (1)

Background, (2) Research Questions, (3) Search

terms, (4) Selection criteria, (5) Quality checklist and

procedures, (6) Data extraction strategy, and (7) Data

synthesis strategy. The research stated search string.

The search string was (hybrid learning AND (smart

OR intelligent)) AND (recommend* OR

classificat*) AND (systems). Figure 1 is described

about the studies selection strategy. Inclusion and

exclusion criteria were used to select the main study.

The results of the article from these criteria will be

iCAST-ES 2021 - International Conference on Applied Science and Technology on Engineering Science

182

reviewed by the researcher. The inclusion and

exclusion criteria can be seen in Table 1.

Figure 1: Studies Selection Strategy.

Table 1: The Inclusion and Exclusion Criteria.

Name Description

Inclusion

Studies in academic and industry using

large- and small-scale data sets

Studies discussing and comparing modeling

performance in the area of smart

recommendation s

stems

For studies that have both the conference and

journal versions, only the journal version

will be include

For duplicate publications of the same study,

only the most complete and newest one will

e include

Exclusion

Studies without a strong validation or

including experimental results of smart

recommendation s

ste

Studies not written in English

Table 2, 3, 4 are shown the filtering process result.

The research started with 4931 articles then after third

layer filtering, there found 50 articles became the data

of the SLR as shown in Table 5. The parameters used

to extract the data include: (1) feature datasets, (2)

hybrid deep learning method, (3) programming

languages, and (4) evaluation parameters.

Table 2: Filtering Process Result of Step Retrieve Initial

List Result.

Keywords ACM

IEEE

Explore

Science

Direct

Springer

Lin

hybrid deep

learning smart

recommend*

systems

1.716 15 152 615

hybrid deep

learning smart

classificat*

stems

1.719 47 137 530

Table 3: Filtering Process Result of Step Exclude Based on

Title and Abstract.

Keywords ACM

IEEE

Explore

Science

Direct

Springer

Lin

hybrid deep

learning smart

recommend*

systems

23 5 2 19

hybrid deep

learning smart

classificat*

stems

16 14 8 7

Table 4: Filtering Process Result of Step Exclude Based on

Full Text.

Keywords ACM

IEEE

lore

Science

Direct

Springer

Lin

hybrid deep

learning smart

recommend*

systems

12 4 1 9

hybrid deep

learning smart

classificat*

stems

4 11 5 5

Table 5: Filtering Process Recapitulation.

Phase Amount

Retrieve Initial List Result 4.931

Exclude Based On Title and Abstract Result 94

Exclude Based On Full Text Result 50

3 RESULTS AND DISCUSSIONS

3.1 The Most Used Datasets

The analysis results of datasets that are widely used

can be seen in Figure 2. Figure 2 shows that the most

widely used dataset in hybrid deep learning smart

A Systematic Literature Review on Hybrid Deep Learning Smart Recommendation Systems

183

recommendation system research was textual dataset,

which is 34 out of 50 total articles.

The research with textual dataset are (Wang &

Cao, 2011), (Liu, Zhang, & Gulla, 2021), (Feng, Li,

Ge, Luo, & Ng, 2021), (Jamal, Xianqiao, Al-

Turjman, & Ullah, A Deep Learning–based Approach

for Emotions Classification in Big Corpus of

Imbalanced Tweets, 2021), (Aliannejadi, Zamani,

Crestani, & Croft, 2021), (Cheng, Shen, Huang, &

Zhu, 2021), (Kimmel, Brack, & Marshall, 2021),

(Jawarneh, et al., 2020), (Vijayalakshmi,

Vinayakamurthy, & Anuradha, 2020), (Gunjal,

Yadav, & Kshirsagar, 2020), (Zou, Gu, Song, Liu, &

Yao, 2017), (Chiu, Huang, Gupta, & Akman, 2021),

(Benlamri & Zhang, 2014), (Motwani, A., Shukla,

P.K., & Pawar, M., 2021), (Qadir, Ever, & Batunlu,

2021), (Anthony Jnr, 2021), (Li, Li, Zhang, Zhong, &

Cheng, 2019), (Vora & Rajamani, 2019), (Torres-

Ruiz, et al., 2020), (Jelodar, et al., 2021) (Karo,

Ramdhani, Ramadhelza, & Aufa, 2020), (Ahmad

Mahmud & Azuana Ramli, 2020), (Zhu, Wang,

Zhong, Li, & Sheng, 2021), (Popoola, Adebisi,

Hammoudeh, Gui, & Gacanin, 2021), (Liang, Zhu,

Zhang, Cheng, & Jin, 2020), (Jahangir, et al., 2021),

(Cuzzocrea, et al., 2020), (Mlika & Karoui, 2020),

(Ibrahim, Saleh, Elgaml, & Abdelsalam, 2020),

(Serano, 2020), (Fang, et al., 2021), (Shafqat, et al.,

2021), and (Masud, et al., 2021).

Next is the image dataset totalling 11 articles,

namely (Yue, et al., 2021), (Cui, Yu, Wu, Liu, &

Wang, 2021), (Khosroshahi, Razavi, Sangar, &

Majidzadeh, 2021), (Luo, Yang, Tang, & Zhang,

2020), (Rafi & Akthar, 2021), (Joshi & Sharma,

2021), (Su & Wei, 2020), (Anjna, Sood, & Singh,

2020), (Khan, Nazir, García-Magariño, & Hussain,

2021), (Louati, 2020), and (Kristiansen, et al., 2021).

Figure 2: The Result of Research Question 1.

Only 1 article from (Mohammed, Elhoseny,

Abdulkareem, Mostafa, & Maashi, 2021) used an

audio-based dataset, as well as a video-based dataset.

There are also studies that use two types of datasets.

Research from (Wang, et al., 2021) used textual and

video. Research from (Li, et al., 2021) and

(Chintamani, Kumar, & Karan, 2021) used image and

audio datasets.

3.2 The Most Used Methods

The analysis results of the most used methods can be

seen in Figure 3. The most widely used hybrid deep

learning method is a combination of the

Convolutional Neural Network (CNN) and Long

Short-Term Memory (LSTM) which are the

development of the Recurrent Neural Network

(RNN) method. LSTM and CNN are also widely

combined with other methods in building a hybrid

deep learning smart recommendation system, but the

amount is not as many as the combination of CNN

and LSTM.

(Wang, et al., 2021) on their research leveraged

hybrid deep learning to distill the textual contents for

more distinguishable features. The research by (Yue,

et al., 2021) used the hybrid deep learning model to

extract discriminative spatial features (CNN) and to

encode temporal information from the encrypted

image sequences (leveraged LSTM).

Figure 3: The Result of Research Question 2.

The combination of CNN and LSTM was also

conducted by (Khan, Nazir, García-Magariño, &

Hussain, 2021), they used CNN for the classification

of spatial data while LSTM for temporal data. The

same with (Liang, Zhu, Zhang, Cheng, & Jin, 2020),

(Feng, Li, Ge, Luo, & Ng, 2021) and (Rafi & Akthar,

2021) they used CNN_LSTM for classification

phase. (Louati, 2020) also used CNN for

classification and CNN-LSTM for traffic prediction.

Different with (Kimmel, Brack, & Marshall,

2021), they utilized CNN and RNN to analyze cell

motility data. RNN modified with CNN became RNN

autoencoders and the research stated that it capable of

learning motility features in an unsupervised manner

and capturing variation between myogenic cells in the

latent space.

1111

Textual Image Audio Video Textual

and

Image

Textual

and

Video

Image

and

Audio

Articles Amount

0 5 10 15 20 25

CNN and LSTM

LSTM with others…

CNN with others…

MLP with others

Others

Articles Amount

iCAST-ES 2021 - International Conference on Applied Science and Technology on Engineering Science

184

3.3 The Most Proposed Programming

Language

Based on the analysis of 50 articles, there are 9 papers

that use python as the programming language and 3

combined python with other languange. The research

was conducted by (Liu, Zhang, & Gulla, 2021), (Su

& Wei, 2020), (Kristiansen, et al., 2021), (Joshi &

Sharma, 2021), (Yue, et al., 2021), (Rafi & Akthar,

2021), (Liang, Zhu, Zhang, Cheng, & Jin, 2020),

(Khosroshahi, Razavi, Sangar, & Majidzadeh, 2021),

(Chiu, Huang, Gupta, & Akman, 2021), (Popoola,

Adebisi, Hammoudeh, Gui, & Gacanin, 2021),

(Motwani, A., Shukla, P.K., & Pawar, M., 2021), and

(Jawarneh, et al., 2020). Python is widely used

because it has modules that strongly support machine

learning, such as NumPy, TensorFlow, Keras,

Pandas, PyTorch, Matplotlib, Scikit-learn. Based on

the data in Figure 4, there are also Matlab, R, and PHP

programming language that are proposed for a hybrid

learning smart recommendation system.

Figure 4: The Result of Research Question 3.

3.4 The Most Used Evaluation

Parameters

There are several parameters used in evaluating the

performance of the hybrid deep learning smart

recommendation system. The analysis result showed

that accuracy, precision, recall, and F1-Score

parameters are most often used as evaluation

parameters in smart recommendation systems.

4 CONCLUSIONS

The research was a Systematic Literature Review

(SLR) research. The research started with 4931

articles from digital libraries namely ACM Digital

Library, IEEE Explore, ScienceDirect and also

SpringerLink. After third layer filtering, there found

50 articles became the data of the SLR. It can be

conluded that the analysis result, namely:

1. The most widely used dataset in hybrid deep

learning smart recommendation system

research was textual dataset.

2. The most widely used hybrid deep learning

method was a combination of the

Convolutional Neural Network (CNN) and

Long Short-Term Memory (LSTM)

3. Python is widely used because it has modules

that strongly support machine learning, such as

NumPy, TensorFlow, Keras, Pandas, PyTorch,

Matplotlib, Scikit-learn.

4. Accuracy, precision, recall, and F1-Score

parameters are most often used as evaluation

parameters in smart recommendation systems.

ACKNOWLEDGEMENTS

Gratitude is dedicated to Politeknik Negeri Bali who

has funded this researchand support for this research.

REFERENCES

Adiyansjah, Gunawan, A., & Suhartono, D. (2019). Music

Recommender System Based on Genre using

Convolutional Recurrent Neural Networks. Procedia

Computer Science, 99-109.

Ahmad Mahmud, N. F., & Azuana Ramli, N. (2020).

Hybrid Classification Method to Detect the Presence of

Human in a Smart Building Environment. 2020

International Conference on Data Analytics for

Business and Industry: Way Towards a Sustainable

Economy (ICDABI) (pp. 1-5). Sakheer, Bahrain: IEEE.

Aliannejadi, M., Zamani, H., Crestani, F., & Croft, W. B.

(2021). Context-aware Target Apps Selection and

Recommendation for Enhancing Personal Mobile

Assistants. ACM Transactions on Information Systems,

1-30.

Anjna, Sood, M., & Singh, K. P. (2020). Hybrid System for

Detection and Classification of Plant Disease Using

Qualitative Texture Features Analysis. Procedia

Computer Science, 1056-1065.

Anthony Jnr, B. (2021). A Case-based Reasoning

Recommender System for Sustainable Smart City

Development. AI & SOCIETY, 159-183.

Benabderrahmane, S., Mellouli, N., & Lamolle, M. (2018).

On the Predictive Analysis of Behavioral Massive Job

Data Using Embedded Clustering and Deep Recurrent

Neural Networks. Knowledge-Based Systems, 955-113.

Benlamri, R., & Zhang, X. (2014). Context-aware

Recommender for Mobile Learners. Human-centric

Computing and Information Sciences.

0 10203040

Python

Python and others

matlab

Matlab and others

PHP

Not Mentioned

Articles Amount

A Systematic Literature Review on Hybrid Deep Learning Smart Recommendation Systems

185

Brusaferria, A., Matteuccib, M., Spinellia, S., & Vitali, A.

(2020). Learning Behavioral Models by Recurrent

Neural Networks Withdiscrete Latent Representations

With Application to a Flexible Ndustrial Conveyor.

Computers in Industry, 103263.

Cheng, W., Shen, Y., Huang, L., & Zhu, Y. (2021). Dual-

Embedding Based Deep Latent Factor Models for

Recommendation. ACM Transactions on Knowledge

Discovery from Data, 1-24.

Chintamani, R. D., Kumar, P., & Karan, R. (2021). Motor

Imagery Classification Based on Hybrid Feature

Extraction and Deep Neural Network. 2021

International Conference on Artificial Intelligence and

Smart Systems (ICAIS) (pp. 885-893). Coimbatore,

India: IEEE.

Chiu, M.-C., Huang, J.-H., Gupta, S., & Akman, G. (2021).

Developing a personalized recommendation system in

a smart product service system based on unsupervised

learning model. Computers in Industry, 103421.

Cui, Z., Yu, F., Wu, S., Liu, Q., & Wang, L. (2021).

Disentangled Item Representation for Recommender

Systems. ACM Transactions on Intelligent Systems and

Technology, 1-20.

Cuzzocrea, A., Leung, C. K., Deng, D., Mai, J., Jiang, F., &

Fadda, E. (2020). A Combined Deep-Learning and

Transfer-Learning Approach for Supporting Social

Influence Prediction. Procedia Computer Science, 170-

177.

Dewi, K., & Harjoko, A. (2010). Kid's song classification

based on mood parameters using K-Nearest Neighbor

classification method and Self Organizing Map. 2010

International Conference on Distributed Frameworks

for Multimedia Applications, (pp. 1-5). Yogyakarta.

Djellali, C., & Adda, M. (2020). A New Hybrid Deep

Learning Model based-Recommender System using

Artificial Neural Network and Hidden Markov Model.

Procedia Computer Science, 214-220.

Fang, M., Chen, Y., Xue, R., Wang, H., Chakraborty, N.,

Su, T., & Dai, Y. (2021). A hybrid machine learning

approach for hypertension risk prediction. Neural

Computing and Applications.

Feng, Y., Li, C., Ge, J., Luo, B., & Ng, V. (2021).

Recommending Statutes: A Portable Method Based on

Neural Networks. ACM Transactions on Knowledge

Discovery from Data, 1-22.

Gunjal, S., Yadav, S., & Kshirsagar, D. B. (2020). A hybrid

scalable collaborative filtering based recommendation

system using ontology and incremental SVD algorithm.

2020 International Conference on Smart Innovations in

Design, Environment, Management, Planning and

Computing (ICSIDEMPC) (pp. 39-45). Aurangabad,

India: IEEE.

Ibrahim, T., Saleh, A., Elgaml, N., & Abdelsalam, M.

(2020). A Fog Based Recommendation System for

Promoting the Performance of E-learning

Environments. Computers & Electrical Engineering,

106791.

Jahangir, H., Tayarani, H., Gougheri, S. S., Golkar, M. A.,

Ahmadian, A., & Elkamel, A. (2021). Deep Learning-

Based Forecasting Approach in Smart Grids With

Microclustering and Bidirectional LSTM Network.

IEEE Transactions on Industrial Electronics

, 8298-

8309.

Jain, D., Mahanti, A., Shamsolmoali, P., & Manikandan, R.

(2020). Deep Neural Learning Techniques With Long

Short-term Memory for Gesture Recognition. Neural

Computing and Applications, 16073-16089.

Jamal, N., Xianqiao, C., Al-Turjman, F., & Ullah, F.

(2021). A Deep Learning–based Approach for

Emotions Classification in Big Corpus of Imbalanced

Tweets. ACM Transactions on Asian and Low-

Resource Language Information Processing, 1-16.

Jamal, N., Xianqiao, C., Al-Turjman, F., & Ullah, F.

(2021). A Deep Learning–based Approach for

Emotions Classification in Big Corpus of Imbalanced

Tweets. ACM Transactions on Asian and Low-

Resource Language Information Processing, 1-16.

Jawarneh, I. M., Bellavista, P., Corradi, A., Foschini, L.,

Montanari, R., Berrocal, J., & Murillo, J. M. (2020). A

Pre-Filtering Approach for Incorporating Contextual

Information Into Deep Learning Based Recommender

Systems. IEEE Access, 40485-40498.

Jelodar, H., Wang, Y., Xiao, G., Rabbani, M., Zhao, R.,

Ayobi, S., . . . Masood, I. (2021). Recommendation

System Based on Semantic Scholar Mining and Topic

Modeling on Conference Publications. Soft Computing,

3675-3696.

Jha, S., Prashar, D., Long, H., & Taniar, D. (2020).

Recurrent neural network for detecting malware.

computers & security, 102037.

Joshi, A., & Sharma, K. (2021). Hybrid Topology of Graph

Convolution and Autoencoder Deep Network For

Multiple Sclerosis Lesion Segmentation. 2021

International Conference on Artificial Intelligence and

Smart Systems (ICAIS) (pp. 1529-1534). Coimbatore,

India: IEEE.

Karo, I. M., Ramdhani, R., Ramadhelza, A. W., & Aufa, B.

Z. (2020). A Hybrid Classification Based on Machine

Learning Classifiers to Predict Smart Indonesia

Program. 2020 Third International Conference on

Vocational Education and Electrical Engineering

(ICVEE) (pp. 1-5). Surabaya, Indonesia: IEEE.

Khan, S., Nazir, S., García-Magariño, I., & Hussain, A.

(2021). Deep Learning-based Urban Big Data Fusion in

Smart Cities: Towards Traffic Monitoring and Flow-

preserving Fusion. Computers & Electrical

Engineering, 106906.

Khosroshahi, S., Razavi, S., Sangar, A., & Majidzadeh, K.

(2021). Deep Neural Networks-based Offline Writer

Identification Using Heterogeneous Handwriting Data:

an Evaluation via a Novel Standard Dataset. Journal of

Ambient Intelligence and Humanized Computing.

Kimmel, J. C., Brack, A. S., & Marshall, W. F. (2021).

Deep Convolutional and Recurrent Neural Networks

for Cell Motility Discrimination and Prediction.

IEEE/ACM Transactions on Computational Biology

and Bioinformatics, 562-574.

Kitchenham, & Charters. (2007). Guidelines in performing

Systematic Literature Reviews in Software

Engineering. EBSE Technical Report version 2.3.

iCAST-ES 2021 - International Conference on Applied Science and Technology on Engineering Science

186

Koehn, D., Lessmann, S., & Schaal, M. (2020). Predicting

Online Shopping Behaviour From Clickstream Data

Using Deep Learning. Expert Systems With

Applications, 113342.

Kristiansen, S., Nikolaidis, K., Plagemann, T., Goebel, V.,

Traaen, G. M., Overland, B., . . . Bendz, C. (2021).

Machine Learning for Sleep Apnea Detection with

Unattended Sleep Monitoring at Home. ACM

Transactions on Computing for Healthcare, 1-25.

Li, H., Li, H., Zhang, S., Zhong, Z., & Cheng, J. (2019).

Intelligent Learning System Based on Personalized

Recommendation Technology. Neural Computing and

Applications, 4455-4462.

Li, H., Pun, C.-M., Xu, F., Pan, L., Zong, R., Gao, H., &

Lu, H. (2021). A Hybrid Feature Selection Algorithm

Based on a Discrete Artificial Bee Colony for

Parkinson's Diagnosis. ACM Transactions on Internet

Technology, 1-22.

Liang, S., Zhu, B., Zhang, Y., Cheng, S., & Jin, J. (2020).

A Double Channel CNN-LSTM Model for Text

Classification. 2020 IEEE 22nd International

Conference on High Performance Computing and

Communications; IEEE 18th International Conference

on Smart City; IEEE 6th International Conference on

Data Science and Systems (HPCC/SmartCity/DSS) (pp.

1316-1321). Yanuca Island, Cuvu, Fiji: IEEE.

Liu, P., Zhang, L., & Gulla, J. A. (2021). Multilingual

Review-aware Deep Recommender System via Aspect-

based Sentiment Analysis. ACM Transactions on

Information Systems, 1-33.

Louati, A. (2020). A Hybridization of Deep Learning

Techniques to Predict and Control Traffic

Disturbances. Artificial Intelligence Review, 5675-

5704.

Luo, L., Yang, Z.-X., Tang, L., & Zhang, K. (2020). An

ELM-Embedded Deep Learning Based Intelligent

Recognition System for Computer Numeric Control

Machine Tools. IEEE Access, 24616-24629.

Maree, M., Rattrout, A., Altawil, M., & Belkhatir, M.

(2021). Multi-modality Search and Recommendation

on Palestinian Cultural Heritage Based on the Holy-

Land Ontology and Extrinsic Semantic Resources.

Journal on Computing and Cultural Heritage, 1-23.

Masud, M., Singh, P., Gaba, G. S., Kaur, A., Alghamdi, R.

A., Alrashoud, M., & Alqahtani, S. A. (2021).

CROWD: Crow Search and Deep Learning based

Feature Extractor for Classification of Parkinson’s

Disease. ACM Transactions on Internet Technology, 1-

18.

Mlika, F., & Karoui, W. (2020). Proposed Model to

Intelligent Recommendation System based on Markov

Chains and Grouping of Genres. Procedia Computer

Science, 868-877.

Mohammed, M. A., Elhoseny, M., Abdulkareem, K. H.,

Mostafa, S. A., & Maashi, M. S. (2021). A Multi-agent

Feature Selection and Hybrid Classification Model for

Parkinson's Disease Diagnosis. ACM Transactions on

Multimedia Computing, 1-22.

Motwani, A., Shukla, P.K., & Pawar, M. (2021). Novel

Framework Based on Deep Learning and Cloud

Analytics for Smart Patient Monitoring and

Recommendation (Spmr). Journal of Ambient

Intelligence and Humanized Computing.

Nilashi, M., Bagherifard, K., Rahmani, M., & Rafe, V.

(2017). A Recommender System for Tourism Industry

Using Cluster Ensemble and Prediction Machine

Learning Techniques. Computers & Industrial

Engineering, 357-368.

Nilashi, M., Ibrahim, O., & Ithnin, N. (2014). Hybrid

Recommendation Approaches for Multi-criteria

Collaborative Filtering. Expert Systems with

Applications, 3879-3900.

Popoola, S. I., Adebisi, B., Hammoudeh, M., Gui, G., &

Gacanin, H. (2021). Hybrid Deep Learning for Botnet

Attack Detection in the Internet-of-Things Networks.

IEEE Internet of Things Journal, 4944-4956.

Qadir, Z., Ever, E., & Batunlu, C. (2021). Use of Neural

Network Based Prediction Algorithms for Powering Up

Smart Portable Accessories. Neural Processing Letters,

721-756.

Rafi, S. M., & Akthar, S. (2021). ECG Classification using

a Hybrid Deeplearning Approach. 2021 International

Conference on Artificial Intelligence and Smart

Systems (ICAIS) (pp. 302-305). Coimbatore, India:

IEEE.

Serano, W. (2020). Genetic and Deep Learning Clusters

Based on Neural Networks for Management Decision

Structures. Neural Computing and Applications, 4187-

4211.

Shafqat, S., Fayyaz, M., Khattak, H., Bilal, M., Khan, S.,

Ishtiaq, O., . . . Chatterjee, P. (2021). Leveraging Deep

Learning for Designing Healthcare Analytics Heuristic

for Diagnostics. Neural Processing Letters.

Snyder, H. (2019). Literature Review as a Research

Methodology: an Overview and Guidelines. Journal of

Business Research, 333-339.

Su, C., & Wei, J. (2020). Hybrid Model of Vehicle

Recognition Based on Convolutional Neural Network.

2020 IEEE 22nd International Conference on High

Performance Computing and Communications; IEEE

18th International Conference on Smart City; IEEE 6th

International Conference on Data Science and Systems

(HPCC/SmartCity/DSS) (pp. 1246-1251). Yanuca

Island, Cuvu, Fiji: IEEE.

Torres-Ruiz, M., Mata, F., Zagal, R., Guzmán, G.,

Quintero, R., & Moreno-Ibarra , M. (2020). A

Recommender System to Generate Museum Itineraries

Applying Augmented Reality and Social-sensor

Mining Techniques. Virtual Reality, 175-189.

Turkut, U., Tuncer, A., Savran, H., & Yılmaz, S. (2020).

An Online Recommendation System Using Deep

Learning for Textile Products. 2020 International

Congress on Human-Computer Interaction,

Optimization and Robotic Applications (HORA), (pp. 1-

4). Ankara, Turkey.

Vijayalakshmi, K., Vinayakamurthy, M., & Anuradha.

(2020). A Hybrid Recommender System using

MultiClassifier Regression Model for Autism

Detection. 2020 International Conference on Smart

A Systematic Literature Review on Hybrid Deep Learning Smart Recommendation Systems

187

Technologies in Computing, Electrical and Electronics

(ICSTCEE) (pp. 139-144). Bengaluru, India: IEEE.

Vora, D., & Rajamani, K. (2019). A Hybrid Classification

Model for Prediction of Academic Performance of

Students: a Big Data Application. Evolutionary

Intelligence.

Wang, K., Zhang, T., Xue, T., Lu, Y., & Na, S. (2019). E-

Commerce Personalized Recommendation Analysis by

Deeply-learned Clustering. J. Vis. Commun. Image R.

Wang, W., & Cao, L. (2011). Interactive Sequential Basket

Recommendation by Learning Basket Couplings and

Positive/Negative Feedback. ACM Transactions on

Information Systems, 1-26.

Wang, W., Duan, L.-Y., Jiang, H., Jing, P., Song, X., & Nie,

L. (2021). Market2Dish: Health-Aware Food

Recommendation. ACM Transactions on Multimedia

Computing, 1-19.

Wu, B., & Ye, Y. (2020). BSPR: Basket-Sensitive

Personalized Ranking for Product Recommendation.

Information Sciences.

Xu, Y., Yang, Y., Han, J., Wang, E., Ming, J., & Xiong, H.

(2019). Slanderous user detection with modified

recurrent neural networks in recommender system.

Information Sciences, 265-281.

Xu, Y., Yang, Y., Han, J., Wang, E., Ming, J., & Xiong, H.

(2019). Slanderous User Detection with Modified

Recurrent Neural Networks in Recommender System.

Information Sciences, 265-281.

Yue, Z., Ding, S., Zhao, L., Zhang, Y., Cao, Z., Tanveer,

M., . . . Zheng, X. (2021). Privacy-preserving Time-

series Medical Images Analysis Using a Hybrid Deep

Learning Framework. ACM Transactions on Internet

Technology, 1-21.

Zhang, M., & Bockstedt, J. (2020). Complements and

Substitutes in Online Product Recommendations: The

Differential Effects on Consumers’ Willingness to Pay.

Information and Management.

Zhu, H., Wang, L., Zhong, S., Li, Y., & Sheng, V. S. (2021).

A Hybrid Deep Network Framework for Android

Malware Detection. IEEE Transactions on Knowledge

and Data Engineering, 1-1.

Zou, L., Gu, Y., Song, J., Liu, W., & Yao, Y. (2017).

HDNN-CF: A hybrid deep neural networks

collaborative filtering architecture for event

recommendation. 2017 IEEE SmartWorld, Ubiquitous

Intelligence Computing, Advanced Trusted Computed,

Scalable Computing Communications, Cloud Big Data

Computing, Internet of People and Smart City

Innovation (SmartWorld/SCALCOM/UIC/ATC/CBD

Com/IOP/SCI) (pp. 1-8). San Francisco, CA, USA:

IEEE.

iCAST-ES 2021 - International Conference on Applied Science and Technology on Engineering Science

188