Matching Items (2)
Filtering by
- All Subjects: time series
- Genre: Academic theses
- Creators: Adame, Bradley
Description
The amount of time series data generated is increasing due to the integration of sensor technologies with everyday applications, such as gesture recognition, energy optimization, health care, video surveillance. The use of multiple sensors simultaneously
for capturing different aspects of the real world attributes has also led to an increase in dimensionality from uni-variate to multi-variate time series. This has facilitated richer data representation but also has necessitated algorithms determining similarity between two multi-variate time series for search and analysis.
Various algorithms have been extended from uni-variate to multi-variate case, such as multi-variate versions of Euclidean distance, edit distance, dynamic time warping. However, it has not been studied how these algorithms account for asynchronous in time series. Human gestures, for example, exhibit asynchrony in their patterns as different subjects perform the same gesture with varying movements in their patterns at different speeds. In this thesis, we propose several algorithms (some of which also leverage metadata describing the relationships among the variates). In particular, we present several techniques that leverage the contextual relationships among the variates when measuring multi-variate time series similarities. Based on the way correlation is leveraged, various weighing mechanisms have been proposed that determine the importance of a dimension for discriminating between the time series as giving the same weight to each dimension can led to misclassification. We next study the robustness of the considered techniques against different temporal asynchronies, including shifts and stretching.
Exhaustive experiments were carried on datasets with multiple types and amounts of temporal asynchronies. It has been observed that accuracy of algorithms that rely on data to discover variate relationships can be low under the presence of temporal asynchrony, whereas in case of algorithms that rely on external metadata, robustness against asynchronous distortions tends to be stronger. Specifically, algorithms using external metadata have better classification accuracy and cluster separation than existing state-of-the-art work, such as EROS, PCA, and naive dynamic time warping.
for capturing different aspects of the real world attributes has also led to an increase in dimensionality from uni-variate to multi-variate time series. This has facilitated richer data representation but also has necessitated algorithms determining similarity between two multi-variate time series for search and analysis.
Various algorithms have been extended from uni-variate to multi-variate case, such as multi-variate versions of Euclidean distance, edit distance, dynamic time warping. However, it has not been studied how these algorithms account for asynchronous in time series. Human gestures, for example, exhibit asynchrony in their patterns as different subjects perform the same gesture with varying movements in their patterns at different speeds. In this thesis, we propose several algorithms (some of which also leverage metadata describing the relationships among the variates). In particular, we present several techniques that leverage the contextual relationships among the variates when measuring multi-variate time series similarities. Based on the way correlation is leveraged, various weighing mechanisms have been proposed that determine the importance of a dimension for discriminating between the time series as giving the same weight to each dimension can led to misclassification. We next study the robustness of the considered techniques against different temporal asynchronies, including shifts and stretching.
Exhaustive experiments were carried on datasets with multiple types and amounts of temporal asynchronies. It has been observed that accuracy of algorithms that rely on data to discover variate relationships can be low under the presence of temporal asynchrony, whereas in case of algorithms that rely on external metadata, robustness against asynchronous distortions tends to be stronger. Specifically, algorithms using external metadata have better classification accuracy and cluster separation than existing state-of-the-art work, such as EROS, PCA, and naive dynamic time warping.
ContributorsGarg, Yash (Author) / Candan, Kasim Selcuk (Thesis advisor) / Chowell-Punete, Gerardo (Committee member) / Tong, Hanghang (Committee member) / Davulcu, Hasan (Committee member) / Sapino, Maria Luisa (Committee member) / Arizona State University (Publisher)
Created2015
Description
Speculation regarding interstate conflict is of great concern to many, if not, all people. As such, forecasting interstate conflict has been an interest to experts, scholars, government officials, and concerned citizens. Presently, there are two approaches to the problem of conflict forecasting with divergent results. The first tends to use a bird’s eye view with big data to forecast actions while missing the intimate details of the groups it is studying. The other opts for more grounded details of cultural meaning and interpretation, yet struggles in the realm of practical application for forecasting. While outlining issues with both approaches, an important question surfaced: are actions causing interpretations and/or are the interpretations driving actions? In response, the Theory of Narrative Conflict (TNC) is proposed to begin answering these questions. To properly address the complexity of forecasting and of culture, TNC draws from a number of different sources, including narrative theory, systems theory, nationalism, and the expression of these in strategic communication.
As a case study, this dissertation examines positions of both the U.S. and China in the South and East China Seas over five years. Methodologically, this dissertation demonstrates the benefit of content analysis to identify local narratives and both stabilizing and destabilizing events contained in thousands of news articles over a five-year period. Additionally, the use of time series and a Markov analysis both demonstrate usefulness in forecasting. Theoretically, TNC displays the usefulness of narrative theory to forecast both actions driven by narrative and common interpretations after events.
Practically, this dissertation demonstrates that current efforts in the U.S. and China have not resulted in an increased understanding of the other country. Neither media giant demonstrates the capacity to be critical of their own national identity and preferred interpretation of world affairs. In short, the battle for the hearts and minds of foreign persons should be challenged.
As a case study, this dissertation examines positions of both the U.S. and China in the South and East China Seas over five years. Methodologically, this dissertation demonstrates the benefit of content analysis to identify local narratives and both stabilizing and destabilizing events contained in thousands of news articles over a five-year period. Additionally, the use of time series and a Markov analysis both demonstrate usefulness in forecasting. Theoretically, TNC displays the usefulness of narrative theory to forecast both actions driven by narrative and common interpretations after events.
Practically, this dissertation demonstrates that current efforts in the U.S. and China have not resulted in an increased understanding of the other country. Neither media giant demonstrates the capacity to be critical of their own national identity and preferred interpretation of world affairs. In short, the battle for the hearts and minds of foreign persons should be challenged.
ContributorsNolen, Matthew Scott (Author) / Corman, Steven R. (Thesis advisor) / Adame, Bradley (Committee member) / Simon, Denis (Committee member) / Arizona State University (Publisher)
Created2017