Search results
1 – 10 of 102Khaled Hamed Alyoubi, Fahd Saleh Alotaibi, Akhil Kumar, Vishal Gupta and Akashdeep Sharma
The purpose of this paper is to describe a new approach to sentence representation learning leading to text classification using Bidirectional Encoder Representations from…
Abstract
Purpose
The purpose of this paper is to describe a new approach to sentence representation learning leading to text classification using Bidirectional Encoder Representations from Transformers (BERT) embeddings. This work proposes a novel BERT-convolutional neural network (CNN)-based model for sentence representation learning and text classification. The proposed model can be used by industries that work in the area of classification of similarity scores between the texts and sentiments and opinion analysis.
Design/methodology/approach
The approach developed is based on the use of the BERT model to provide distinct features from its transformer encoder layers to the CNNs to achieve multi-layer feature fusion. To achieve multi-layer feature fusion, the distinct feature vectors of the last three layers of the BERT are passed to three separate CNN layers to generate a rich feature representation that can be used for extracting the keywords in the sentences. For sentence representation learning and text classification, the proposed model is trained and tested on the Stanford Sentiment Treebank-2 (SST-2) data set for sentiment analysis and the Quora Question Pair (QQP) data set for sentence classification. To obtain benchmark results, a selective training approach has been applied with the proposed model.
Findings
On the SST-2 data set, the proposed model achieved an accuracy of 92.90%, whereas, on the QQP data set, it achieved an accuracy of 91.51%. For other evaluation metrics such as precision, recall and F1 Score, the results obtained are overwhelming. The results with the proposed model are 1.17%–1.2% better as compared to the original BERT model on the SST-2 and QQP data sets.
Originality/value
The novelty of the proposed model lies in the multi-layer feature fusion between the last three layers of the BERT model with CNN layers and the selective training approach based on gated pruning to achieve benchmark results.
Details
Keywords
Gangting Huang, Qichen Wu, Youbiao Su, Yunfei Li and Shilin Xie
In order to improve the computation efficiency of the four-point rainflow algorithm, a new fast four-point rainflow cycle counting algorithm (FFRA) using a novel loop iteration…
Abstract
Purpose
In order to improve the computation efficiency of the four-point rainflow algorithm, a new fast four-point rainflow cycle counting algorithm (FFRA) using a novel loop iteration mode is proposed.
Design/methodology/approach
In this new algorithm, the loop iteration mode is simplified by reducing the number of iterations, tests and deletions. The high efficiency of the new algorithm makes it a preferable candidate in fatigue life online estimation of structural health monitoring systems.
Findings
The extensive simulation results show that the extracted cycles by the new FFRA are the same as those by the four-point rainflow cycle counting algorithm (FRA) and the three-point rainflow cycle counting algorithm (TRA). Especially, the simulation results indicate that the computation efficiency of the FFRA has improved an average of 12.4 times compared to the FRA and an average of 8.9 times compared to the TRA. Moreover, the equivalence of cycle extraction results between the FFRA and the FRA is proved mathematically by utilizing some fundamental properties of the rainflow algorithm. Theoretical proof of the efficiency improvement of the FFRA in comparison to the FRA is also given.
Originality/value
This merit makes the FFRA preferable in online monitoring systems of structures where fatigue life estimation needs to be accomplished online based on massive measured data. It is noticeable that the high efficiency of the FFRA attributed to the simple loop iteration, which provides beneficial guidance to improve the efficiency of existing algorithms.
Details
Keywords
Oscar F. Bustinza, Luis M. Molina Fernandez and Marlene Mendoza Macías
Machine learning (ML) analytical tools are increasingly being considered as an alternative quantitative methodology in management research. This paper proposes a new approach for…
Abstract
Purpose
Machine learning (ML) analytical tools are increasingly being considered as an alternative quantitative methodology in management research. This paper proposes a new approach for uncovering the antecedents behind product and product–service innovation (PSI).
Design/methodology/approach
The ML approach is novel in the field of innovation antecedents at the country level. A sample of the Equatorian National Survey on Technology and Innovation, consisting of more than 6,000 firms, is used to rank the antecedents of innovation.
Findings
The analysis reveals that the antecedents of product and PSI are distinct, yet rooted in the principles of open innovation and competitive priorities.
Research limitations/implications
The analysis is based on a sample of Equatorian firms with the objective of showing how ML techniques are suitable for testing the antecedents of innovation in any other context.
Originality/value
The novel ML approach, in contrast to traditional quantitative analysis of the topic, can consider the full set of antecedent interactions to each of the innovations analyzed.
Details
Keywords
Tingting Tian, Hongjian Shi, Ruhui Ma and Yuan Liu
For privacy protection, federated learning based on data separation allows machine learning models to be trained on remote devices or in isolated data devices. However, due to the…
Abstract
Purpose
For privacy protection, federated learning based on data separation allows machine learning models to be trained on remote devices or in isolated data devices. However, due to the limited resources such as bandwidth and power of local devices, communication in federated learning can be much slower than in local computing. This study aims to improve communication efficiency by reducing the number of communication rounds and the size of information transmitted in each round.
Design/methodology/approach
This paper allows each user node to perform multiple local trainings, then upload the local model parameters to a central server. The central server updates the global model parameters by weighted averaging the parameter information. Based on this aggregation, user nodes first cluster the parameter information to be uploaded and then replace each value with the mean value of its cluster. Considering the asymmetry of the federated learning framework, adaptively select the optimal number of clusters required to compress the model information.
Findings
While maintaining the loss convergence rate similar to that of federated averaging, the test accuracy did not decrease significantly.
Originality/value
By compressing uplink traffic, the work can improve communication efficiency on dynamic networks with limited resources.
Details
Keywords
Jasleen Kaur and Khushdeep Dharni
The stock market generates massive databases of various financial companies that are highly volatile and complex. To forecast daily stock values of these companies, investors…
Abstract
Purpose
The stock market generates massive databases of various financial companies that are highly volatile and complex. To forecast daily stock values of these companies, investors frequently use technical analysis or fundamental analysis. Data mining techniques coupled with fundamental and technical analysis types have the potential to give satisfactory results for stock market prediction. In the current paper, an effort is made to investigate the accuracy of stock market predictions by using the combined approach of variables from technical and fundamental analysis for the creation of a data mining predictive model.
Design/methodology/approach
We chose 381 companies from the National Stock Exchange of India's CNX 500 index and conducted a two-stage data analysis. The first stage is identifying key fundamental variables and constructing a portfolio based on that study. Artificial neural network (ANN), support vector machines (SVM) and decision tree J48 were used to build the models. The second stage entails applying technical analysis to forecast price movements in the companies included in the portfolios. ANN and SVM techniques were used to create predictive models for all companies in the portfolios. We also estimated returns using trading decisions based on the model's output and then compared them to buy-and-hold returns and the return of the NIFTY 50 index, which served as a benchmark.
Findings
The results show that the returns of both the portfolios are higher than the benchmark buy-and-hold strategy return. It can be concluded that data mining techniques give better results, irrespective of the type of stock, and have the ability to make up for poor stocks. The comparison of returns of portfolios with the return of NIFTY as a benchmark also indicates that both the portfolios are generating higher returns as compared to the return generated by NIFTY.
Originality/value
As stock prices are influenced by both technical and fundamental indicators, the current paper explored the combined effect of technical analysis and fundamental analysis variables for Indian stock market prediction. Further, the results obtained by individual analysis have also been compared. The proposed method under study can also be utilized to determine whether to hold stocks for the long or short term using trend-based research.
Details
Keywords
Mathias Chukwudi Isiani, Ngozika Anthonia Obi-Ani, Chikelue Chris Akabuike, Stanley Jachike Onyemechalu, Sochima P. Okafor and Sopuluchukwu Amarachukwu Dimelu
The overall aim of this research is to interpret Ikenga and Ofo creativity as it is revered in Igbo societies. Igbo creativity, especially interpreted through material culture…
Abstract
Purpose
The overall aim of this research is to interpret Ikenga and Ofo creativity as it is revered in Igbo societies. Igbo creativity, especially interpreted through material culture, suffers the threat of extinction resulting from the forces of modernity. Forces of modernisation, which appear in the personae of Christianity, education, urbanisation and industrialisation, denigrated indigenous creativity, brandishing them as devious, fetish and primitive. Ironically, in most cases, the drivers of such narratives keep these “fetish” items in their museums and will give a lot to preserve them.
Design/methodology/approach
This study centred mostly on several communities in the Nsukka area of Igboland, Nigeria. It relied on both primary and secondary sources of historical enquiry. This qualitative research discussed the nuances of the subject matter as it relates to Igbo cosmos. These approaches involved visiting the study area and conducting personal interviews.
Findings
Archaeologists do often rely on material culture to study, periodise and date past human societies. In this study, it is found that material culture, an expression of indigenous creativity, best interprets how society survived or related with their environment. This paper examined two Igbo sculpted artefacts – Ikenga and Ofo – while unearthing the intricacies in Igbo cosmology as regards creativity, spirituality and society.
Originality/value
The shapes, motifs, patterns and designs depict an imaginary history, the intellectualism of the past and even the present. This serves as an objective alternative to the twisted colonial narrative on Igbo material culture and consequently contribute to ongoing efforts to preserve, protect and promote cultural heritage resources in this part of the world.
Details
Keywords
Rucha Wadapurkar, Sanket Bapat, Rupali Mahajan and Renu Vyas
Ovarian cancer (OC) is the most common type of gynecologic cancer in the world with a high rate of mortality. Due to manifestation of generic symptoms and absence of specific…
Abstract
Purpose
Ovarian cancer (OC) is the most common type of gynecologic cancer in the world with a high rate of mortality. Due to manifestation of generic symptoms and absence of specific biomarkers, OC is usually diagnosed at a late stage. Machine learning models can be employed to predict driver genes implicated in causative mutations.
Design/methodology/approach
In the present study, a comprehensive next generation sequencing (NGS) analysis of whole exome sequences of 47 OC patients was carried out to identify clinically significant mutations. Nine functional features of 708 mutations identified were input into a machine learning classification model by employing the eXtreme Gradient Boosting (XGBoost) classifier method for prediction of OC driver genes.
Findings
The XGBoost classifier model yielded a classification accuracy of 0.946, which was superior to that obtained by other classifiers such as decision tree, Naive Bayes, random forest and support vector machine. Further, an interaction network was generated to identify and establish correlations with cancer-associated pathways and gene ontology data.
Originality/value
The final results revealed 12 putative candidate cancer driver genes, namely LAMA3, LAMC3, COL6A1, COL5A1, COL2A1, UGT1A1, BDNF, ANK1, WNT10A, FZD4, PLEKHG5 and CYP2C9, that may have implications in clinical diagnosis.
Details
Keywords
Serena Summa, Alex Mircoli, Domenico Potena, Giulia Ulpiani, Claudia Diamantini and Costanzo Di Perna
Nearly 75% of EU buildings are not energy-efficient enough to meet the international climate goals, which triggers the need to develop sustainable construction techniques with…
Abstract
Purpose
Nearly 75% of EU buildings are not energy-efficient enough to meet the international climate goals, which triggers the need to develop sustainable construction techniques with high degree of resilience against climate change. In this context, a promising construction technique is represented by ventilated façades (VFs). This paper aims to propose three different VFs and the authors define a novel machine learning-based approach to evaluate and predict their energy performance under different boundary conditions, without the need for expensive on-site experimentations
Design/methodology/approach
The approach is based on the use of machine learning algorithms for the evaluation of different VF configurations and allows for the prediction of the temperatures in the cavities and of the heat fluxes. The authors trained different regression algorithms and obtained low prediction errors, in particular for temperatures. The authors used such models to simulate the thermo-physical behavior of the VFs and determined the most energy-efficient design variant.
Findings
The authors found that regression trees allow for an accurate simulation of the thermal behavior of VFs. The authors also studied feature weights to determine the most relevant thermo-physical parameters. Finally, the authors determined the best design variant and the optimal air velocity in the cavity.
Originality/value
This study is unique in four main aspects: the thermo-dynamic analysis is performed under different thermal masses, positions of the cavity and geometries; the VFs are mated with a controlled ventilation system, used to parameterize the thermodynamic behavior under stepwise variations of the air inflow; temperatures and heat fluxes are predicted through machine learning models; the best configuration is determined through simulations, with no onerous in situ experimentations needed.
Details
Keywords
Wenhao Zhou and Hailin Li
This study aims to propose a combined effect framework to explore the relationship between research and development (R&D) team networks, knowledge diversity and breakthrough…
Abstract
Purpose
This study aims to propose a combined effect framework to explore the relationship between research and development (R&D) team networks, knowledge diversity and breakthrough technological innovation. In contrast to conventional linear net effects, the article explores three possible types of team configuration within enterprises and their breakthrough innovation-driving mechanisms based on machine learning methods.
Design/methodology/approach
Based on the patent application data of 2,337 Chinese companies in the biopharmaceutical manufacturing industry to construct the R&D team network, the study uses the K-Means method to explore the configuration types of R&D teams with the principle of greatest intergroup differences. Further, a decision tree model (DT) is utilized to excavate the conditional combined relationships between diverse team network configuration factors, knowledge diversity and breakthrough innovation. The network driving mechanism of corporate breakthrough innovation is analyzed from the perspective of team configurations.
Findings
It has been discerned that in the biopharmaceutical manufacturing industry, there exist three main types of enterprise R&D team configurations: tight collaboration, knowledge expansion and scale orientation, which reflect the three resource investment preferences of enterprises in technological innovation, network relationships, knowledge resources and human capital. The results highlight both the crowding-out effects and complementary effects between knowledge diversity and team network characteristics in tight collaborative teams. Low knowledge diversity and high team structure holes (SHs) are found to be the optimal team configuration conditions for breakthrough innovation in knowledge-expanding and scale-oriented teams.
Originality/value
Previous studies have mainly focused on the relationship between the external collaboration network and corporate innovation. Moreover, traditional regression methods mainly describe the linear net effects between variables, neglecting that technological breakthroughs are a comprehensive concept that requires the combined action of multiple factors. To address the gap, this article proposes a combination effect framework between R&D teams and enterprise breakthrough innovation, further improving social network theory and expanding the applicability of data mining methods in the field of innovation management.
Details
Keywords
When ownership starts getting dispersed among several individuals, families, branches, and generations, a need for organizing communications and decision-making usually arises to…
Abstract
When ownership starts getting dispersed among several individuals, families, branches, and generations, a need for organizing communications and decision-making usually arises to ensure functional relationships within the family. The need for a shared vision and mutually agreed ways of handling the shared ownership emerges, and a process for developing a family governance structure is often initiated. Family governance, hence, appears to be a central topic in family business research, but we still lack a more profound and specific understanding of how the owner family uses different family governance mechanisms to manage specific situations with possible conflicting goals, interests, and opinions, or just to develop the shared ownership further for or together with the next generation. The aim of this chapter is to give an overview and highlight different processes developed by the family within owner families with dispersed ownership to identify and align governance goals. This overview intends to broaden the understanding of what the role of family governance, as a family internal mechanism, can be in owner families with dispersed ownership among several family members.
Details