Julia

Voulgaris, Zacharias

doi:10.1007/978-3-319-63962-8_268-2

Cited by 2 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, sufficient parameters are critical to capture nuances and prevent under-fitting issues. With the use of regularization, poor generalization can often be avoided (Voulgaris, 2016). The regularization of models uses parameters differently by merging them while simultaneously fine-tuning them.…”

Section: Forecasting Modelsmentioning

confidence: 99%

A segmented machine learning modeling approach of social media for predicting occupancy

Ampountolas

Legg

2021

IJCHM

View full text Add to dashboard Cite

Purpose This study aims to predict hotel demand through text analysis by investigating keyword series to increase demand predictions’ precision. To do so, this paper presents a framework for modeling hotel demand that incorporates machine learning techniques. Design/methodology/approach The empirical forecasting is conducted by introducing a segmented machine learning approach of leveraging hierarchical clustering tied to machine learning and deep learning techniques. These features allow the model to yield more precise estimates. This study evaluates an extensive range of social media–derived words with the most significant probability of gradually establishing an understanding of an optimal outcome. Analyzes were performed on a major hotel chain in an urban market setting within the USA. Findings The findings indicate that while traditional methods, being the naïve approach and ARIMA models, struggled with forecasting accuracy, segmented boosting methods (XGBoost) leveraging social media predict hotel occupancy with greater precision for all examined time horizons. Additionally, the segmented learning approach improved the forecasts’ stability and robustness while mitigating common overfitting issues within a highly dimensional data set. Research limitations/implications Incorporating social media into a segmented learning framework can augment the current generation of forecasting methods’ accuracy. Moreover, the segmented learning approach mitigates the negative effects of market shifts (e.g. COVID-19) that can reduce in-production forecasts’ life-cycles. The ability to be more robust to market deviations will allow hospitality firms to minimize development time. Originality/value The results are expected to generate insights by providing revenue managers with an instrument for predicting demand.

show abstract

Section: Forecasting Modelsmentioning

confidence: 99%

A segmented machine learning modeling approach of social media for predicting occupancy

Ampountolas

Legg

2021

IJCHM

View full text Add to dashboard Cite

show abstract

“…XGBoost is based on a gradient algorithm that uses scalable, regularized tree boosting to provide flexible and optimized regression and classification trees (Chen and Guestrin, 2016;Hastie et al, 2001a, b). Through regularization, poor generalization within models' predictive outcomes can be mitigated (Voulgaris, 2016), while inputting a sizable amount of independent factors can minimize under-fitting issues. Additionally, incorporating broader scale matrixes into XGBoost models allows for capturing nuances within independent factors and contribution attributes (Ampountolas and Legg, 2021).…”

Section: Gradient Boostingmentioning

confidence: 99%

Endings, freezing, and new beginnings: the return of customer comfort to Massachusetts restaurants following the pandemic

Palabiyik

Cronin

Markham-Bagnera

et al. 2022

JHTI

View full text Add to dashboard Cite

PurposeThis study investigates restaurant patrons' comfort level with the sudden shift in the dining-in climate within the state of Massachusetts during the onset of the COVID-19 pandemic.Design/methodology/approachAn exploratory study utilized learning algorithms via gradient boosting techniques on surveyed restaurant patrons to identify which restaurant operational attributes and patron demographics predict in-dining comfort levels.FindingsPast consumers' eating habits determine how much their behavior will change during a pandemic. However, their dining-in frequency is not a predictor of their post-pandemic dining-in outlook. The individuals who were more comfortable dining in prior to the pandemic dined in more often during the COVID pandemic. However, they had a poorer outlook on when dining in would return to normal. Although there are no clear indicators of when and how customers will embrace the new norm (a combination of pre-, peri-, and post-pandemic), the results show that some innovative approaches, such as limiting service offerings, are not well accepted by customers.Practical implicationsThe study offers several managerial implications for foodservice providers (i.e. restaurants, delivery services, pick-up) and investors. In particular, the study provides insights into the cognitive factors that determine diners' behavioral change in response to a pandemic and their comfort level. Operators must pay attention to these factors and consider different offering strategies when preparing to operate their business amid a pandemic.Originality/valueThis is a study of a specific location and period. It was conducted in Massachusetts before a vaccine was available. The restaurant industry was beset with uncertainty. It fills a gap in the current literature focused on the COVID-19 pandemic in customers' transition from pre-COVID-19 dining-in behaviors to customers' refreshed COVID-19 outlook and industry compliance with newly established hygiene and safety standards.

show abstract

Julia

Cited by 2 publications

References 0 publications

A segmented machine learning modeling approach of social media for predicting occupancy

A segmented machine learning modeling approach of social media for predicting occupancy

Endings, freezing, and new beginnings: the return of customer comfort to Massachusetts restaurants following the pandemic

Contact Info

Product

Resources

About