Limits...
Lightweight Adaptation of Classifiers to Users and Contexts: Trends of the Emerging Domain.

Vildjiounaite E, Gimel'farb G, Kyllönen V, Peltola J - ScientificWorldJournal (2015)

Bottom Line: Such applications have to adapt to a new user or unknown context at runtime just from interaction with the user, preferably in fairly lightweight ways, that is, requiring limited user effort to collect training data and limited time of performing the adaptation.This paper analyses adaptation trends in several emerging domains and outlines promising ideas, proposed for making multimodal classifiers user-specific and context-specific without significant user efforts, detailed domain knowledge, and/or complete retraining of the classifiers.Based on this analysis, this paper identifies important application characteristics and presents guidelines to consider these characteristics in adaptation design.

View Article: PubMed Central - PubMed

Affiliation: VTT Technical Research Centre of Finland, Kaitoväylä 1, 90571 Oulu, Finland.

ABSTRACT
Intelligent computer applications need to adapt their behaviour to contexts and users, but conventional classifier adaptation methods require long data collection and/or training times. Therefore classifier adaptation is often performed as follows: at design time application developers define typical usage contexts and provide reasoning models for each of these contexts, and then at runtime an appropriate model is selected from available ones. Typically, definition of usage contexts and reasoning models heavily relies on domain knowledge. However, in practice many applications are used in so diverse situations that no developer can predict them all and collect for each situation adequate training and test databases. Such applications have to adapt to a new user or unknown context at runtime just from interaction with the user, preferably in fairly lightweight ways, that is, requiring limited user effort to collect training data and limited time of performing the adaptation. This paper analyses adaptation trends in several emerging domains and outlines promising ideas, proposed for making multimodal classifiers user-specific and context-specific without significant user efforts, detailed domain knowledge, and/or complete retraining of the classifiers. Based on this analysis, this paper identifies important application characteristics and presents guidelines to consider these characteristics in adaptation design.

No MeSH data available.


Modelling situation-dependency scenario; dashed lines denote optional data.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4581555&req=5

fig1: Modelling situation-dependency scenario; dashed lines denote optional data.

Mentions: The lightweight runtime adaptation to previously unseen contexts typically employs a mixed model in Figure 1: the fine descriptors (“context cues” in Figure 1) or their types are predefined at design time, whereas the higher-level ones (the “situations” in Figure 1) are defined dynamically at runtime. Dynamic definition of high-level contexts can be achieved via analysis of primary data, for example, segmentation [34] or matrix factorisation [35]. The context change can be also detected via analysis of external factors, for example, context features or user interaction (e.g., users may explicitly declare the change by naming a new context or implicitly indicate it by correcting classification errors and requesting adaptation). Below we assume that analysis of external factors was employed.


Lightweight Adaptation of Classifiers to Users and Contexts: Trends of the Emerging Domain.

Vildjiounaite E, Gimel'farb G, Kyllönen V, Peltola J - ScientificWorldJournal (2015)

Modelling situation-dependency scenario; dashed lines denote optional data.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4581555&req=5

fig1: Modelling situation-dependency scenario; dashed lines denote optional data.
Mentions: The lightweight runtime adaptation to previously unseen contexts typically employs a mixed model in Figure 1: the fine descriptors (“context cues” in Figure 1) or their types are predefined at design time, whereas the higher-level ones (the “situations” in Figure 1) are defined dynamically at runtime. Dynamic definition of high-level contexts can be achieved via analysis of primary data, for example, segmentation [34] or matrix factorisation [35]. The context change can be also detected via analysis of external factors, for example, context features or user interaction (e.g., users may explicitly declare the change by naming a new context or implicitly indicate it by correcting classification errors and requesting adaptation). Below we assume that analysis of external factors was employed.

Bottom Line: Such applications have to adapt to a new user or unknown context at runtime just from interaction with the user, preferably in fairly lightweight ways, that is, requiring limited user effort to collect training data and limited time of performing the adaptation.This paper analyses adaptation trends in several emerging domains and outlines promising ideas, proposed for making multimodal classifiers user-specific and context-specific without significant user efforts, detailed domain knowledge, and/or complete retraining of the classifiers.Based on this analysis, this paper identifies important application characteristics and presents guidelines to consider these characteristics in adaptation design.

View Article: PubMed Central - PubMed

Affiliation: VTT Technical Research Centre of Finland, Kaitoväylä 1, 90571 Oulu, Finland.

ABSTRACT
Intelligent computer applications need to adapt their behaviour to contexts and users, but conventional classifier adaptation methods require long data collection and/or training times. Therefore classifier adaptation is often performed as follows: at design time application developers define typical usage contexts and provide reasoning models for each of these contexts, and then at runtime an appropriate model is selected from available ones. Typically, definition of usage contexts and reasoning models heavily relies on domain knowledge. However, in practice many applications are used in so diverse situations that no developer can predict them all and collect for each situation adequate training and test databases. Such applications have to adapt to a new user or unknown context at runtime just from interaction with the user, preferably in fairly lightweight ways, that is, requiring limited user effort to collect training data and limited time of performing the adaptation. This paper analyses adaptation trends in several emerging domains and outlines promising ideas, proposed for making multimodal classifiers user-specific and context-specific without significant user efforts, detailed domain knowledge, and/or complete retraining of the classifiers. Based on this analysis, this paper identifies important application characteristics and presents guidelines to consider these characteristics in adaptation design.

No MeSH data available.