Sales forecasting by combining clustering and machine-learning techniques for computer retailing

Chen, I-Fei; Lu, Chi-Jie

doi:10.1007/s00521-016-2215-x

Cited by 52 publications

(32 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Forecasting is an important tool in a number of fields, including weather and climate [ 1 – 3 ], agriculture [ 4 , 5 ], air quality [ 6 , 7 ], and consumer activity [ 8 – 10 ]. When operationalized for use in real time, predictions from probabilistic forecasts can be used in decision-making to inform, for example, emergency food aid allocation [ 4 ] or profit maximization [ 8 ].…”

Section: Introductionmentioning

confidence: 99%

Development and validation of influenza forecasting for 64 temperate and tropical countries

Krämer

Shaman

2019

PLoS Comput Biol

View full text Add to dashboard Cite

Accurate forecasts of influenza incidence can be used to inform medical and public health decision-making and response efforts. However, forecasting systems are uncommon in most countries, with a few notable exceptions. Here we use publicly available data from the World Health Organization to generate retrospective forecasts of influenza peak timing and peak intensity for 64 countries, including 18 tropical and subtropical countries. We find that accurate and well-calibrated forecasts can be generated for countries in temperate regions, with peak timing and intensity accuracy exceeding 50% at four and two weeks prior to the predicted epidemic peak, respectively. Forecasts are significantly less accurate in the tropics and subtropics for both peak timing and intensity. This work indicates that, in temperate regions around the world, forecasts can be generated with sufficient lead time to prepare for upcoming outbreak peak incidence.

show abstract

Section: Introductionmentioning

confidence: 99%

Development and validation of influenza forecasting for 64 temperate and tropical countries

Krämer

Shaman

2019

PLoS Comput Biol

View full text Add to dashboard Cite

show abstract

“…Numerical forecasting—the computational real-time generation of calibrated predictions on time scales allowing application and validation—has a long history of use in the fields of weather and climate 1 – 3 . In recent decades, numerical forecasts have been developed for and applied to a number of new industries and disciplines, including agriculture 4 , 5 , air quality 6 , 7 , consumer activity 8 – 10 , fiscal policy 11 , and political elections 12 . These forecasts allow stakeholders to prepare for predicted future events and to respond accordingly.…”

Section: Introductionmentioning

confidence: 99%

“…For example, forecasts of crop yields help governments decide whether food must be imported to meet population needs, and inform decisions concerning the receipt of emergency food aid 5 . Meanwhile, many companies use sales forecasting when deciding how much of a product to stock in order to maximize profits 8 . In public health, forecasting methods have been developed using mathematical models and Bayesian inference methods and used to predict the growth and spread of infectious diseases such as influenza 13 – 18 , dengue 19 – 21 , Ebola 22 – 24 , and, most recently, Zika 25 , 26 .…”

Section: Introductionmentioning

confidence: 99%

Assessing the Use of Influenza Forecasts and Epidemiological Modeling in Public Health Decision Making in the United States

Doms

Krämer

Shaman

2018

Sci Rep

View full text Add to dashboard Cite

Although forecasts and other mathematical models have the potential to play an important role in mitigating the impact of infectious disease outbreaks, the extent to which these tools are used in public health decision making in the United States is unclear. Throughout 2015, we invited public health practitioners belonging to three national public health organizations to complete a cross-sectional survey containing questions on model awareness, model use, and communication with modelers. Of 39 respondents, 46.15% used models in their work, and 20.51% reported direct communication with those who create models. Over half (64.10%) were aware that influenza forecasts exist. The need for improved communication between practitioners and modelers was overwhelmingly endorsed, with over 50% of participants indicating the need for models more relevant to public health questions, increased frequency of telecommunication, and more plain language in discussing models. Model use for public health decision making must be improved if models are to reach their full potential as public health tools. Increased quality and frequency of communication between practitioners and modelers could be particularly useful in achieving this goal. It is important that improvements be made now, rather than waiting for the next public health crisis to occur.

show abstract

“…According to Chen and Lu [20], clustering can improve the performance of classification models. Therefore, k-means clustering was performed with the aim of grouping together customers with similar purchasing patterns, into a number of k pre-specified clusters.…”

Section: Clusteringmentioning

confidence: 99%

Shortening Delivery Times by Predicting Customers’ Online Purchases: A Case Study in the Fashion Industry

Weingarten

Spinler

2020

Information Systems Management

View full text Add to dashboard Cite

Online retailers still struggle with the disadvantage of delivery times compared to traditional brick and mortar stores. With the emergence of big data analytics, it has become possible to extract meaningful knowledge from the volume of data that online retailers collect on their website. Nevertheless, limited research exists that investigates how this data can be used to optimize delivery times for customers. The goal of this paper is to develop a prediction model for anticipatory shipping, which predicts customers' online purchases with the aim of shipping products in advance, and subsequently minimize delivery times. Different forecasting methods in combination with k-means clustering are applied to test if, and how early, it is possible to predict online purchases. Results indicate that customer purchases are, to a certain extent, predictable, but anticipatory shipping comes at a high cost due to wrongly sent products. The proposed prediction model can easily be implemented and used to predict purchases, which can also be leveraged for other areas of application besides anticipatory shipping.

show abstract

Sales forecasting by combining clustering and machine-learning techniques for computer retailing

Cited by 52 publications

References 44 publications

Development and validation of influenza forecasting for 64 temperate and tropical countries

Development and validation of influenza forecasting for 64 temperate and tropical countries

Assessing the Use of Influenza Forecasts and Epidemiological Modeling in Public Health Decision Making in the United States

Shortening Delivery Times by Predicting Customers’ Online Purchases: A Case Study in the Fashion Industry

Contact Info

Product

Resources

About