Bayesian approach to sensor-based context awareness

Panu Korpipää (Corresponding Author), Miika Koskinen, Johannes Peltola, Satu-Marja Mäkelä, Tapio Seppänen

    Research output: Contribution to journalArticleScientificpeer-review

    90 Citations (Scopus)

    Abstract

    The usability of a mobile device and services can be enhanced by context awareness. The aim of this experiment was to expand the set of generally recognizable constituents of context concerning personal mobile device usage. Naive Bayesian networks were applied to classify the contexts of a mobile device user in her normal daily activities. The distinguishing feature of this experiment in comparison to earlier context recognition research is the use of a naive Bayes framework, and an extensive set of audio features derived partly from the algorithms of the upcoming MPEG-7 standard. The classification was based mainly on audio features measured in a home scenario. The classification results indicate that with a resolution of one second in segments of 5–30 seconds, situations can be extracted fairly well, but most of the contexts are likely to be valid only in a restricted scenario. Naive Bayes framework is feasible for context recognition. In real world conditions, the recognition accuracy using leave-one-out cross validation was 87% of true positives and 95% of true negatives, averaged over nine eight-minute scenarios containing 17 segments of different lengths and nine different contexts. Respectively, the reference accuracies measured by testing with training data were 88% and 95%, suggesting that the model was capable of covering the variability introduced in the data on purpose. Reference recognition accuracy in controlled conditions was 96% and 100%, respectively. However, from the applicability viewpoint, generalization remains a problem, as from a wider perspective almost any feature may refer to many possible real world situations.
    Original languageEnglish
    Pages (from-to)113-124
    Number of pages12
    JournalPersonal and Ubiquitous Computing
    Volume7
    Issue number2
    DOIs
    Publication statusPublished - 2003
    MoE publication typeA1 Journal article-refereed

    Fingerprint

    Mobile devices
    Sensors
    Motion Picture Experts Group standards
    Bayesian networks
    Experiments
    Testing
    Bayesian approach
    Context-awareness
    Sensor
    Scenarios
    Experiment

    Keywords

    • audio context
    • bayesian networks
    • context awareness
    • context recognition
    • mobile computing
    • mobile devices
    • mobile services
    • sensors

    Cite this

    Korpipää, Panu ; Koskinen, Miika ; Peltola, Johannes ; Mäkelä, Satu-Marja ; Seppänen, Tapio. / Bayesian approach to sensor-based context awareness. In: Personal and Ubiquitous Computing. 2003 ; Vol. 7, No. 2. pp. 113-124.
    @article{d85c10da1f9147cdbbae30ec14ad41b1,
    title = "Bayesian approach to sensor-based context awareness",
    abstract = "The usability of a mobile device and services can be enhanced by context awareness. The aim of this experiment was to expand the set of generally recognizable constituents of context concerning personal mobile device usage. Naive Bayesian networks were applied to classify the contexts of a mobile device user in her normal daily activities. The distinguishing feature of this experiment in comparison to earlier context recognition research is the use of a naive Bayes framework, and an extensive set of audio features derived partly from the algorithms of the upcoming MPEG-7 standard. The classification was based mainly on audio features measured in a home scenario. The classification results indicate that with a resolution of one second in segments of 5–30 seconds, situations can be extracted fairly well, but most of the contexts are likely to be valid only in a restricted scenario. Naive Bayes framework is feasible for context recognition. In real world conditions, the recognition accuracy using leave-one-out cross validation was 87{\%} of true positives and 95{\%} of true negatives, averaged over nine eight-minute scenarios containing 17 segments of different lengths and nine different contexts. Respectively, the reference accuracies measured by testing with training data were 88{\%} and 95{\%}, suggesting that the model was capable of covering the variability introduced in the data on purpose. Reference recognition accuracy in controlled conditions was 96{\%} and 100{\%}, respectively. However, from the applicability viewpoint, generalization remains a problem, as from a wider perspective almost any feature may refer to many possible real world situations.",
    keywords = "audio context, bayesian networks, context awareness, context recognition, mobile computing, mobile devices, mobile services, sensors",
    author = "Panu Korpip{\"a}{\"a} and Miika Koskinen and Johannes Peltola and Satu-Marja M{\"a}kel{\"a} and Tapio Sepp{\"a}nen",
    year = "2003",
    doi = "10.1007/s00779-003-0237-8",
    language = "English",
    volume = "7",
    pages = "113--124",
    journal = "Personal and Ubiquitous Computing",
    issn = "1617-4909",
    publisher = "Springer",
    number = "2",

    }

    Bayesian approach to sensor-based context awareness. / Korpipää, Panu (Corresponding Author); Koskinen, Miika; Peltola, Johannes; Mäkelä, Satu-Marja; Seppänen, Tapio.

    In: Personal and Ubiquitous Computing, Vol. 7, No. 2, 2003, p. 113-124.

    Research output: Contribution to journalArticleScientificpeer-review

    TY - JOUR

    T1 - Bayesian approach to sensor-based context awareness

    AU - Korpipää, Panu

    AU - Koskinen, Miika

    AU - Peltola, Johannes

    AU - Mäkelä, Satu-Marja

    AU - Seppänen, Tapio

    PY - 2003

    Y1 - 2003

    N2 - The usability of a mobile device and services can be enhanced by context awareness. The aim of this experiment was to expand the set of generally recognizable constituents of context concerning personal mobile device usage. Naive Bayesian networks were applied to classify the contexts of a mobile device user in her normal daily activities. The distinguishing feature of this experiment in comparison to earlier context recognition research is the use of a naive Bayes framework, and an extensive set of audio features derived partly from the algorithms of the upcoming MPEG-7 standard. The classification was based mainly on audio features measured in a home scenario. The classification results indicate that with a resolution of one second in segments of 5–30 seconds, situations can be extracted fairly well, but most of the contexts are likely to be valid only in a restricted scenario. Naive Bayes framework is feasible for context recognition. In real world conditions, the recognition accuracy using leave-one-out cross validation was 87% of true positives and 95% of true negatives, averaged over nine eight-minute scenarios containing 17 segments of different lengths and nine different contexts. Respectively, the reference accuracies measured by testing with training data were 88% and 95%, suggesting that the model was capable of covering the variability introduced in the data on purpose. Reference recognition accuracy in controlled conditions was 96% and 100%, respectively. However, from the applicability viewpoint, generalization remains a problem, as from a wider perspective almost any feature may refer to many possible real world situations.

    AB - The usability of a mobile device and services can be enhanced by context awareness. The aim of this experiment was to expand the set of generally recognizable constituents of context concerning personal mobile device usage. Naive Bayesian networks were applied to classify the contexts of a mobile device user in her normal daily activities. The distinguishing feature of this experiment in comparison to earlier context recognition research is the use of a naive Bayes framework, and an extensive set of audio features derived partly from the algorithms of the upcoming MPEG-7 standard. The classification was based mainly on audio features measured in a home scenario. The classification results indicate that with a resolution of one second in segments of 5–30 seconds, situations can be extracted fairly well, but most of the contexts are likely to be valid only in a restricted scenario. Naive Bayes framework is feasible for context recognition. In real world conditions, the recognition accuracy using leave-one-out cross validation was 87% of true positives and 95% of true negatives, averaged over nine eight-minute scenarios containing 17 segments of different lengths and nine different contexts. Respectively, the reference accuracies measured by testing with training data were 88% and 95%, suggesting that the model was capable of covering the variability introduced in the data on purpose. Reference recognition accuracy in controlled conditions was 96% and 100%, respectively. However, from the applicability viewpoint, generalization remains a problem, as from a wider perspective almost any feature may refer to many possible real world situations.

    KW - audio context

    KW - bayesian networks

    KW - context awareness

    KW - context recognition

    KW - mobile computing

    KW - mobile devices

    KW - mobile services

    KW - sensors

    U2 - 10.1007/s00779-003-0237-8

    DO - 10.1007/s00779-003-0237-8

    M3 - Article

    VL - 7

    SP - 113

    EP - 124

    JO - Personal and Ubiquitous Computing

    JF - Personal and Ubiquitous Computing

    SN - 1617-4909

    IS - 2

    ER -