Towards the development of an explainable e-commerce fake review index: An attribute analytics approach

European Journal of Operational Research, Journal Year: 2024, Volume and Issue: 317(2), P. 382 - 400

Published: March 11, 2024

Instruments of corporate risk and reputation assessment tools are quintessentially developed on structured quantitative data linked to financial ratios macroeconomics. An emerging stream studies has challenged this norm by demonstrating improved model prediction capabilities through unstructured textual data. Fake online consumer reviews pose serious threats a business' competitiveness sales performance, directly impacting revenue, market share, brand even survivability. Research shown that as little three negative can lead potential loss 59.2% customers. Amazon, the largest e-commerce retail platform, hosts over 85,000 small-to-medium-size (SME) retailers (UK), selling fifty percent Amazon products worldwide. Despite Amazon's best efforts, fake growing problem causing reputational damage at scale never seen before. While large corporations better equipped handle these problems more efficiently, SMEs become biggest victims scam tactics. Following principles attribute (AA) responsible (RA) analytics, we present novel hybrid method for indexing enterprise call Review Index (RFRI). The proposed modular approach benefits from combination review metadata semantic topic index derived product reviews. We further apply LIME develop Confidence Score, demonstrates importance explainability openness in contemporary analytics within OR domain. Transparency, simplicity our roadmap offers an attractive entry platform practitioners managers industry.

Language: Английский

---

Evaluation of the efficiency of world airports using WENSLO-ARTASI and Monte-Carlo simulation

Journal of Air Transport Management, Journal Year: 2025, Volume and Issue: 124, P. 102749 - 102749

Published: Jan. 21, 2025

Language: Английский

---

Perceived service quality and electronic word-of-mouth on food delivery services: extended theory of planned behaviour

British Food Journal, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 8, 2025

Purpose This study proposed an extended theory of planned behaviour (TPB), that is, considering behavioural beliefs, normative beliefs and control (beliefs) will influence perceived service quality (PSQ) on food delivery services. PSQ (behavioural intention) electronic word-of-mouth (EWOM) (behaviour). In addition, exogenous variables including information from online ratings consumer groups affect the strength relationship between received EWOM service. Design/methodology/approach aimed to investigate mediation moderation (Online groups) effects TPB for Taiwanese consumers ( n = 823). Findings first found a positive different intention). there is EWOM. Online rating has moderating effect Consumer group Originality/value three stages beliefs-intention-behaviour are reciprocal with two paths, starting offline-to-online in terms generating individual belies PSQ. Next, it can generate power return enhance strengthen

Language: Английский

---

The impact of security and privacy perceptions on cryptocurrency app evaluations by users: A text mining study

Investment Management and Financial Innovations, Journal Year: 2025, Volume and Issue: 22(1), P. 173 - 187

Published: Jan. 28, 2025

This study examines how perceived security and privacy influence user ratings of cryptocurrency applications, which are critical for adoption satisfaction amid the growing popularity blockchain technologies rising concerns over information in online platforms mobile apps. The focuses on applications from Android app market. It used text mining methods to investigate 64 thousand text-based reviews star 140 cryptocurrency-related available Google Play store. Using a partially supervised machine learning approach, this first identified reviewer sentiment related security, then employed ordinal regression analysis examine data reveal relationship between threats, concerns, ratings. found that crypto apps average 3.84 out 55 stars, is higher than Productivity (3.46) but lower FinTech (4.29) Banking (4.25) Ordinal revealed threats negatively impact ratings, while robust measures improve them, with model Pseudo R² 0.25. These results have implications both developers platform managers, offering insights enhancing experiences informing future research endeavors domain. contributes literature by integrating Protection Motivation Theory provides structured framework developing an understanding behavior context

Language: Английский

---

Measuring the impact of healthcare service quality of hospitals on customer satisfaction

Corporate and Business Strategy Review, Journal Year: 2024, Volume and Issue: 5(1, special Issue), P. 336 - 345

Published: Jan. 1, 2024

Various studies have been conducted to measure the level of service provided by an institution or service, including hospitals, in order determine customer satisfaction. The purpose this research is assess role quality services offered a hospital predicting This study was at Bunda Purwokerto Hospital. participants consisted 380 samples. Hypothesis testing using Smart PLS method yielded R-square value 0.413 for satisfaction variable. results H1 revealed favourable influence on outpatient patients original sample produced coefficient 0.130, t statistic 2.163 (> 1.96), and p-value 0.031 (< 0.05). relationship between trust job performance deemed positive significant, as evidenced t-statistic exceeding 1.96. confirms that improvement healthcare shapes patient loyalty. Aspects encompassing quality, such physical environment, customer-oriented atmosphere, responsiveness, effective communication, privacy maintenance, safety, demonstrate correlation with loyalty towards hospital’s image.

Language: Английский

---

Harnessing the Power of LLMs for Service Quality Assessment From User-Generated Content

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 99755 - 99767

Published: Jan. 1, 2024

Adopting Large Language Models (LLMs) creates opportunities for organizations to increase efficiency, particularly in sentiment analysis and information extraction tasks. This study explores the efficiency of LLMs real-world applications, focusing on service quality dimension from user-generated content (UGC). For this purpose, we compare performance two (ChatGPT 3.5 Claude 3) three traditional NLP methods using datasets customer reviews (one English one Persian). The results indicate that can achieve notable accuracy (76% ChatGPT 68% (substantial agreement with human raters moderate 3), demonstrating an improvement compared other AI models. However, challenges persist, including discrepancies between model predictions judgments limitations extracting specific dimensions unstructured text. Whereas streamline SQ assessment process, supervision remains essential ensure reliability.

Language: Английский

---

Developing a Framework for Online Review-based Healthcare Service Quality Assessment: A Text-Mining Study (Preprint)

Published: Sept. 5, 2024

The assessment of medical service quality is difficult, due to the information asymmetry between patients and doctors. With development online healthcare platforms, nowadays have more channels reach assess others reviews as a indicator. However, what aspects in we should care that may affect demand doctors remain largely unexplored. goal this study establish comprehensive evaluation system reflect patients' actual experience satisfaction from broader perspective provide scientific basis for improving services. We adopt text mining approach, conducting topic analysis, theory-driven selection, aggregation develop five-dimensional framework based on reviews. This encompasses doctors' expertise, communication attitude, delivery process, outcomes, empathy. also conduct empirical analysis if patients’ positive or negative (i.e., sentiments) dimensions would matter future demand. heterogeneity caused by disease type rank are explored well. Using data large platform China, find polarity patient opinions five significantly show suffering chronic diseases sensitive these comparing with non-chronic diseases. Doctors’ exhibit significant moderating role among 5 dimensions. offers foundational exploration how prioritize when evaluating services, implications management. Our findings both clinicians practical insights delivering high-quality results can guide managers formulating strategies enhance quality, minimize dissatisfaction, strengthen patient-provider relationships.

Language: Английский

---

الطائرات الهيدروجينية كوسيلة آمنة لتجربة السفر المُستدام: بالتركيز على نظرية الإدراك-التوقع

مجلة کلية السياحة والفنادق جامعة المنصورة, Journal Year: 2024, Volume and Issue: 15(2), P. 43 - 84

Published: June 1, 2024

تُعد الطائرات التي تعمل بالهيدروجين الأخضر أداة محورية قد تُستخدم علي نطاق واسع مستقبلا ًمن قبل صناعة النقل الجوي، لتقليل الانبعاثات الكربونية الناتجة عن استخدام الوقود التقليدي في الطيران. سيسهم ذلك التقليل من حدة الأزمات التى تنتُج التغيرات المناخية؛ وكذلك الحد مستويات التلوث الضوضائي الناتج عملية الإحتراق، ضوء توفير تجارب سفر أكثر إستدامة لتحقيق رؤية مصر 2030. هذا السياق، تمت دراسة تقييمات تكنولوجيا الهيدروجينية الأدبيات السابقة؛ مع لم يتم استكشاف العلاقة بين درجة الرضاء تجاه السفر السابقة بهذه والإدراك البيئي لكيفية تقليل هذه للتلوث الجوي . خلال التأثيرات الحدودية لمواقف المسافرين على متن وتجاربهم الفعلية مرحلتى الإدراك والرضاء المُحتمل نحو التجربة السابقة، يسعى البحث إلى تقديم نظرة شاملة لنظرية الإدراك- التوقع تسويقية، التعرف مدى إدراك لطبيعة بيئيًا، تحديد بالوقود وموقف تجربة متنها والتحقق دور خلق مُستدام، فضلاً الدور التسلسلي لموقف وتجربة الفعلي ورضاءهم السابقة. باستخدام استبيان إلكترونى ، تم جمع 414 استجابة المصريين الذين سبق لهم المُجهزة التابعة لبعض شركات الطيران العالمية شهري نوفمبر وديسمبر 2023. أظهرت النتائج أن المواقف الناشئة والتجارب توسطت تسلسليًا الفجوة المُدركة المُحتمل.

---

Determining investment allocation strategies to improve consumer satisfaction based on a preference learning model

Journal of Retailing and Consumer Services, Journal Year: 2024, Volume and Issue: 82, P. 104140 - 104140

Published: Oct. 30, 2024

Language: Английский

---