Prediction of Users' Response Time in Q&amp;A Communities

Burlutskiy, Nikolay; Fish, Andrew; Alí, Nour; Petridis, Miltos

doi:10.1109/icmla.2015.190

Cited by 7 publications

(13 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They demonstrated that DBN-based algorithms outperformed existing methods, such as a nonlinear Support Vector Machines (SVM) and k-means. A similar advantage of DBN in terms of accuracy was shown in [6] where the authors showed that a DBN-based algorithm outperformed other existing methods such as LR, Decision Trees (DT), k Nearest Neighbours (k-NN), and SVM.…”

Section: Related Worksupporting

confidence: 58%

“…This behaviour includes but is not limited to predicting one's location at a particular time [22], customer preferences [26], user activities at home [20,9], and behaviour on social media [27,6]. As a result, a variety of statistical models have been used for predicting user behaviour.…”

Section: Related Workmentioning

confidence: 99%

“…This task is defined by an input feature set X and a class label set Y where the goal is to match labels from the set Y to instances in the set X. The input feature set X is formed from factors potentially affecting the accuracy of predicting user behaviour which are identified in the literature [25,2,12,1,14,6,3]. Then the class label Y is formulated.…”

Section: Machine Learning For Predicting User Behaviourmentioning

confidence: 99%

“…Recently, many researchers have achieved promising results in using Machine Learning (ML) for predicting user activity at home [20,9], in online forums [6], social media [24], and customer preference [26].…”

Section: Introductionmentioning

confidence: 99%

“…Recently deep learning showed promising results in terms of the accuracy but training such models may take significant time [9]. In contrast, simple linear classifier can provide fast but probably less accurate results [6]. The following question is addressed in this paper:…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

An Investigation on Online Versus Batch Learning in Predicting User Behaviour

Burlutskiy

Petridis

Fish

et al. 2016

Research and Development in Intelligent Systems XXXIII

Self Cite

View full text Add to dashboard Cite

An investigation on how to produce a fast and accurate prediction of user behaviour on the Web is conducted. First, the problem of predicting user behaviour as a classification task is formulated and then the main problems of such real-time predictions are specified: the accuracy and time complexity of the prediction. Second, a method for comparison of online and batch (offline) algorithms used for user behaviour prediction is proposed. Last, the performance of these algorithms using the data from a popular question and answer platform, Stack Overflow, is empirically explored. It is demonstrated that a simple online learning algorithm outperforms state-of-the-art batch algorithms and performs as well as a deep learning algorithm, Deep Belief Networks. The proposed method for comparison of online and offline algorithms as well as the provided experimental evidence can be used for choosing a machine learning set-up for predicting user behaviour on the Web in scenarios where the accuracy and the time performance are of main concern.

show abstract

Section: Related Worksupporting

confidence: 58%