Sequential pattern sampling with norm-based utility

Diop, Lamine; Diop, Cheikh Talibouya; Giacometti, Arnaud; Li, Dominique; Soulet, Arnaud

doi:10.1007/s10115-019-01417-3

Cited by 10 publications

(14 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To consider another measure, Moens and Boley had to design a new method [28]. Considering sequences, Diop et al proposed an approach which guarantees that the probability of sampling a sequential pattern is proportional to its frequency [12,13]. It focuses on the frequency measure only.…”

Section: Heuristic Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Anytime mining of sequential discriminative patterns in labeled sequences

et al. 2020

View full text Add to dashboard Cite

Section: Heuristic Methodsmentioning

confidence: 99%

“…Heuristic methods for sequential data have not yet attracted much attention. Diop et al [13] introduced a sampling approach that draws patterns according to the frequency measure only. In [16], Egho et al, propose a promising method: the sampling method misère can be used for any quality measure when exploiting sequences of events.…”

Section: Introductionmentioning

confidence: 99%

Anytime mining of sequential discriminative patterns in labeled sequences

et al. 2020

View full text Add to dashboard Cite

“…For this reason, it would be more advantageous to directly sample the output space (i.e., pattern space) instead of the input space (i.e., data space). To the best of our knowledge, there are methods for sequential data [9] but not for sampling patterns in a data stream. First, stochastic methods [3] require evaluating the measure m on the entire data for selecting the next state of the random walk.…”

Section: Related Workmentioning

confidence: 99%

“…This evaluation is impossible in a data stream because we do not have all the data observations. Second, multi-step random procedures [5,9] have the advantage of not directly evaluating the measure. They consist in drawing a data observation proportionally to the utility sum of patterns that it contains and then, in drawing a pattern from these patterns proportionally to its utility.…”

Section: Related Workmentioning

confidence: 99%

Reservoir Pattern Sampling in Data Streams

Giacometti

Soulet

2021

Machine Learning and Knowledge Discovery in Databases. Research Track

Self Cite

View full text Add to dashboard Cite

Many applications generate data streams where online analysis needs are essential. In this context, pattern mining is a complex task because it requires access to all data observations. To overcome this problem, the state-of-the-art methods maintain a data sample or a compact data structure retaining only recent information on the main patterns. This paper addresses online pattern discovery in data streams based on pattern sampling techniques. Benefiting from reservoir sampling, we propose a generic algorithm, named ResPat, that uses a limited memory space and that integrates a wide spectrum of temporal biases simulating landmark window, sliding window or exponential damped window. For these three window models, we provide fast damping optimizations and we study their temporal complexity. Experiments show that the performance of ResPat algorithms is particularly good. Finally, we illustrate the interest of our approach with online outlier detection in data streams.

show abstract

“…Once a set of patterns has been extracted, we re-encode the dataset, in the same way as authors in [12], [13]. We create a feature for each pattern, putting a Boolean "1" value if the pattern appears in the sequence, "0" otherwise, as illustrated in Table III.…”

Section: E Dataset Re-encodingmentioning

confidence: 99%

A Behavioral Pattern Mining Approach to Model Player Skills in Rocket League

Mathonat¹,

Boulicaut²,

Kaytoue³

2020

2020 IEEE Conference on Games (CoG)

View full text Add to dashboard Cite

Competitive gaming, or esports, is now wellestablished and brought the game industry in a novel era. It comes with many challenges among which evaluating the level of a player, given the strategies and skills she masters. We are interested in automatically identifying the so called skillshots from game traces of Rocket League, a "soccer with rocketpowered cars" game. From a pure data point of view, each skill execution is unique and standard pattern matching may be insufficient. We propose a non trivial data-centric approach based on pattern mining and supervised learning techniques. We show through an extensive set of experiments that most of Rocket League skillshots can be efficiently detected and used for player modelling. It unveils applications for match making, supporting game commentators and learning systems among others.

show abstract

Sequential pattern sampling with norm-based utility

Cited by 10 publications

References 32 publications

Anytime mining of sequential discriminative patterns in labeled sequences

Anytime mining of sequential discriminative patterns in labeled sequences

Reservoir Pattern Sampling in Data Streams

A Behavioral Pattern Mining Approach to Model Player Skills in Rocket League

Contact Info

Product

Resources

About