Rob Loke, R.E.

Rob Loke, PhD

drs. dr. ir.

20192024

Research activity per year

Personal profile

Research interests

I am currrently assistant professor data science @CMIHVA.nl and have a PhD in computer science, an MA in psychology and an MSc in computer science. My earlier research from before 2017 that is not being mentioned here touched upon pattern recognition, image processing, machine learning, robotics, and adaptive control, to mention a few in arbitrary order. Since my PhD was on 3D visualisation and computer graphics, I cannot wait to further extend my current and recent CMIHVA research activities listed below that are almost all in the text domain, requiring mostly natural language processing related stuff---NLP, to the image and video domains :-)

In my CMIHVA projects, I have been quite active with MSc students in measuring the omnivalent corporate reputation construct via a social listening approach based on natural language processing (NLP) in which social media, human-resource vacancy data and online review data on the public web for various stakeholder groups of companies is being targeted and visualized to explore company insights in datastream contents, opinions, and sentiments. Company stakeholder groups can be customers [3,6,7,9,11], employees [4,5], internal stakeholders [13,14] as well as the general public [10,12] for instance. The developed state of the art processing pipelines that have been developed such as custom machine learning for sentiment classification on the specific aspects assortment, product, service and delivery [3], aspect-based sentiment analysis [6,4,7,9], generic semantic search [5,10,13,17], entity extraction [14] and social media analysis [29] are beneficial to feed some of the top applications (numbered for convenience A to G below) in many sectors and markets like ecommerce that we see in companies around us in the professional field nowadays such as:

  • A---(near real-time) dashboards in retail [9], dashboards that are concerned with sustainability topics in many industry sectors including the energy sector for example [10], or, even more specifically, dashboards in fashion with environmental, social, and corporate governance (ESG) topics for instance [11]; 
  • B---decision support systems dashboards in telecom for example [12]; and
  • C---time series prediction dashboards in collaboration with an external retail company [25]. This is not only relevant in marketing but also in fintech for companies' corporate reputation monitoring and risk management.

Besides the social listening approach [26], a visual listening approach [24] in which relevant online image data on the public web is dealt with is envisioned to be applied soon. Social media network analysis as well. 

Again, together with MSc students, I have been targeting other top applications in ecommerce as well:

  • D---chatbots. In [15], an ontology-based chatbot to improve the interaction between the customer and the company is presented, and, in [16], an investigation into the influence of online reviews on corporate reputation with interactive knowledge graphs. 
  • E---recommender systems. With Florensis BV, an online recommender system was built to assist the customer decision making process.

As well as applications with:

  • F---blockchain. Application of blockchain technology is in general potentially relevant for challenges in fintech, logistics and supply chain management that businesses around us might encounter. In [18], a blockchain-based online review system to enhance corporate reputation has been developed.
  • G---fake review detection as an example of fraud detection that has much potential for companies in current society has been targeted in my CMIHVA corporate reputation project from a general data science approach, i.e. defined as the intersection of computer science/IT, math and statistics and domains/business knowledge [1,2], in [8]. A more specific generative AI approach [20] for fake review detection of online products using generative adversarial learning has been applied in [21] and detection of AI-generated fake reviews for corporate reputation management in [23]. As a spinoff application of fake review detection, quality review classification using generative adversarial learning for the hospitality industry has been targeted in [22].  

Besides sophisticated tool development, phenomenon investigation is targeted as well. For instance, mixed method research on customer decision making in specific sectors such as the hospitality sector [19]. Some other examples are in the world of b corp certification [27] and sports [28].

In my current work with CMIHVA, I have also been active with companies directly in industry. For instance, to mention one example, with KPMG Germany, the demand of helicopter spare parts with machine learning methods on logistics and warehouse management data was predicted for ADAC Heliservice.

Besides mentioned datastreams, marketplaces such as bol.com and amazon have been targeted as well. In [21], fake review scores of different companies have been measured on amazon.nl based on product reviews. Similar or other relevant project work is planned to be done on the bol.com marketplace.

Specific processing pipelines need commonly be quite focused to be either data or model driven; however, it is expected that in systems architectural terms a hybrid mixed approach will need to be developed in near future to due advantages.

While most of the applications are operated on desktop laptops and PCs, some should be operated in cloud infrastructure [4]. Edge computing and IOT infrastructure have not been targeted too much yet in the CMIHVA but are beneficial because of foreseen benefits of federated learning in technology transfer on the cloud edge continuum that might boost applications that circumvent data privacy and safety issues in marketing for retailers for instance [30]. Federated learning will technologically link to relevant robotics applications that are foreseen in the CMIHVA.

Via the academic collaboration with Brightdata that I have established within the CMIHVA as one of the first universities worldwide in 2019, we want to stimulate original research proposals for doing good with data (https://brightinitiative.com/). 

References

[1] https://towardsdatascience.com/introduction-to-statistics-e9d72d818745

[2] https://en.wikipedia.org/wiki/Data_science

[3] Sentiment polarity classification of corporate review data with a bidirectional Long-Short Term Memory (biLSTM) neural network architecture https://www.hva.nl/subsites/en/cet/publications/pure-import/sentiment-polarity-classification-of-corporate-review-data-with-a-bidirectional-long-short-term-memory-bilstm-neural-network-architecture.html as well as https://doi.org/10.5220/0009892303100317 (published)

[4] Assessing Corporate Reputation from Online Employee Reviews https://link.springer.com/chapter/10.1007/978-981-16-9268-0_20 (published)

[5] A Company’s Corporate Reputation through the Eyes of Employees Measured with Sentiment Analysis of Online Reviews https://www.scitepress.org/Link.aspx?doi=10.5220/0010620603780385 (published)

[6] Aspect Based Sentiment Analysis on Online Review Data to Predict Corporate Reputation https://www.scitepress.org/Link.aspx?doi=10.5220/0010607203430352 (published)

[7] Exploring Corporate Reputation based on Sentiment Polarities That Are Related to Opinions in Dutch Online Reviews https://www.scitepress.org/PublicationsDetail.aspx?ID=kHC///NTU3o=&t=1 (published)

[8] The Role of Fake Review Detection in Managing Online Corporate Reputation https://www.scitepress.org/PublicationsDetail.aspx?ID=NErJaXyEk4s=&t=1 (published)

[9] Automatic Measurement of Corporate Reputation for Retail Companies from Online Public Data on the Web https://www.hva.nl/subsites/en/cet/publications/pure-import/automatic-measurement-of-corporate-reputation-for-retail-companies-from-online-public-data-on-the-web.html (in prep)

[10] Corporate Reputation of Companies on Twitter Seen from a Sustainability Perspective https://link.springer.com/chapter/10.1007/978-981-16-9272-7_42 (published)

[11] Corporate reputation through the eyes of social responsibility in the fashion industry http://www.icmarktech.org/index.php/en/ (accepted)

[12] Decision Support System for Corporate Reputation based Social Media Listening using a Cross-Source Sentiment Analysis Engine https://www.scitepress.org/PublicationsDetail.aspx?ID=h2uAr7XXd5A=&t=1 (published)

[13] Presence of Corporate Reputation Cues in Company Vacancy Texts Boosts Vacancy Attractiveness as Perceived by Employees https://doi.org/10.5220/0012863400003756 (published) 

[14] Using soft skills to leverage the innovation dimension of Corporate Reputation: A human capital approach (in prep)

[15] An ontology-based chatbot to improve the interaction between the reviewer and the company (accepted)

[16] An Investigation into the Influence of Online Reviews on Corporate Reputation with interactive Knowledge graphs (in prep)

[17] Semantic Search on Aspects of Corporate Reputation (in prep)

[18] The development of a blockchain-based online review system to enhance corporate reputation (in prep)

[19] Exploring the Link between Online Corporate Reputation and Consumer Decision-Making in the Hospitality Industry - A Study of Amsterdam Hotels (in prep)

[20] https://en.wikipedia.org/wiki/Generative_artificial_intelligence

[21] Fake Review Detection of Online Products using Generative Adversarial Learning (in prep)

[22] Quality review classification for the hospitality industry (in prep)

[23] Detecting AI-Generated Reviews for Corporate Reputation Management (in prep)

[24] Liu Liu, Daria Dzyabura, Natalie Mizik (2020) Visual Listening In: Extracting Brand Image Portrayed on Social Media. Marketing Science 39(4):669-686. https://doi.org/10.1287/mksc.2020.1226 

[25] Whisky Auction Price Forecasting: Tasting the Future (in prep)

[26] Westermann, A., Forthmann, J.: Social listening: a potential game changer in reputation management How big data analysis can contribute to understanding stakeholders’ views on organizations. Corp. Commun. Int. J. (2020). ISSN: 1356–3289 https://doi.org/10.1108/CCIJ-01-2020-0028 

[27] Evaluating the impact of B Corp Certification on engagement and satisfaction from the perspective of employees as internal stakeholders (in prep)

[28] The Social Scoreboard: Assessing Social Media’s Impact on Athlete and Sponsored Brands (in prep)

[29] Identifying Brand Advocates: A Neural Network Approach to Social Media Analysis (in prep)

[30] Leveraging Innovative Technology for Enhanced Consumer Choice in Fresh Produce (in prep)

 

Fingerprint

Dive into the research topics where Rob Loke is active. These topic labels come from the works of this person. Together they form a unique fingerprint.
  • 1 Similar Profiles