Margaret Dunham - Academia.edu (original) (raw)
Uploads
Papers by Margaret Dunham
9th International Workshop an Data …, 2010
Profile models based on Hidden Markov Models (HMM) for sequence studies have gained visibility am... more Profile models based on Hidden Markov Models (HMM) for sequence studies have gained visibility among researchers. While the mathematical foundation, the proven algorithms such as Viterbi, Forward and Backward algorithms have certainly provided a rigorous probabilistic platform, the requirement of classic alignment has ensured an extremely high time complexity. We propose the use of another kind of Markov model called Extensible Markov Models (EMM) to create profile architectures that are more efficient in space and time complexity than their HMM counter parts. EMM efficiency comes from an alignment-free paradigm through use of an improved statistical signature form of sequences. The EMM aproach is based on the use sliding p-mers that count every possible p-mer pattern along equal sized segments of a sequence which are then clustered into Markov states. The resulting count vectors shift the position based letter-by-letter sequence analysis problem for phylogenetic trees, classification and search to a more efficient numerical vector space. Using adapted Karlin-Altschul statistics from the Basic Local Alignment Search Tool (BLAST) literature, the EMM based sequence classification also computes a p-value for statistical significance. We present a comparison between profiles generated using profile HMM and EMM.
Journal of The Experimental Analysis of Behavior, 2001
An experiment with rats examined the roles of demarcating stimuli and differential reinforcement ... more An experiment with rats examined the roles of demarcating stimuli and differential reinforcement probability on the development of functional response units. It examined the development of units in a probabilistic, free-operant situation in which the presence of demarcating stimuli was manipulated. In all conditions, behavior became organized into two-response sequences framed by changes in local reinforcement probability. A tone demarcating the beginning and end of contingent response sequences facilitated the development of functional response units, as in chunking, but the same units developed slowly in the absence of the tone. Complex functional response units developed even though reinforcement contingencies remained constant. These findings demonstrate that models of operant learning must include a mechanism for changing the response unit as a function of reinforcement history. Markov models may seem to be a natural technique for modeling response sequences because of their ability to predict individual responses as a function of reinforcement history; however, no class of Markov chain can incorporate changing response units in their predictions.
Workshop on Data Engineering for Wireless and …, 2001
Studies in African linguistics, 2010
This paper presents the Langi verbal system and the various ways in which tense, aspect and mood ... more This paper presents the Langi verbal system and the various ways in which tense, aspect and mood are encoded. Through a description of the structures and uses of the various forms, it attempts to demonstrate how the different conjugations fit together to form a coherent whole, morphologically and semantically, and how in some cases the system has been influenced by surrounding Cushitic languages.
Computer Systems: Science & Engineering, 2005
This paper presents the Langi verbal system and the various ways in which tense, aspect and mood ... more This paper presents the Langi verbal system and the various ways in which tense, aspect and mood are encoded. Through the description of the structures and uses of the various forms, it attempts to demonstrate how the different conjugations fit together to form a coherent whole, morphologically and semantically, and how in some cases the system has been influenced by surrounding Cushitic languages. RESUME Cet article présente le système verbal du langi et les différents moyens mis en oeuvre pour encoder le temps, l'aspect et le mode. A travers la description des structures et emplois des diverses formes, il tente de démontrer comment les conjugaisons diverses forment un système cohérent, sur les plans morphologiques et sémantiques, et comment, dans certains cas, le système a été influencé par les langues couchitiques environnantes. * I thank the following for their helpful comments on earlier drafts of this paper: Christiane Paulian, Zlatka Guentchéva, Denis Creissels, Dave Odden and an anonymous reviewer at SAL. I am also indebted to Derek Nurse and Maarten Mous for pointing out (as well as providing) various articles of interest for this study. 2 This language is relatively unknown to linguistics: when I began studying it in 1996, the only published work dated from 1916 (by Otto Dempwolff). The data presented here is all first hand, and was gathered during fieldwork I carried out in Tanzania during my doctoral studies, the funding for which was provided by the LACITO-CNRS. Oliver Stegen of SIL has started working on the language recently; so far he has presented a paper on the vowel system at CALL (Leiden) in 2000, and has published a paper on derivation (2002). A monograph on Langi is in press: Dunham (forthcoming).
9th International Workshop an Data …, 2010
Profile models based on Hidden Markov Models (HMM) for sequence studies have gained visibility am... more Profile models based on Hidden Markov Models (HMM) for sequence studies have gained visibility among researchers. While the mathematical foundation, the proven algorithms such as Viterbi, Forward and Backward algorithms have certainly provided a rigorous probabilistic platform, the requirement of classic alignment has ensured an extremely high time complexity. We propose the use of another kind of Markov model called Extensible Markov Models (EMM) to create profile architectures that are more efficient in space and time complexity than their HMM counter parts. EMM efficiency comes from an alignment-free paradigm through use of an improved statistical signature form of sequences. The EMM aproach is based on the use sliding p-mers that count every possible p-mer pattern along equal sized segments of a sequence which are then clustered into Markov states. The resulting count vectors shift the position based letter-by-letter sequence analysis problem for phylogenetic trees, classification and search to a more efficient numerical vector space. Using adapted Karlin-Altschul statistics from the Basic Local Alignment Search Tool (BLAST) literature, the EMM based sequence classification also computes a p-value for statistical significance. We present a comparison between profiles generated using profile HMM and EMM.
Journal of The Experimental Analysis of Behavior, 2001
An experiment with rats examined the roles of demarcating stimuli and differential reinforcement ... more An experiment with rats examined the roles of demarcating stimuli and differential reinforcement probability on the development of functional response units. It examined the development of units in a probabilistic, free-operant situation in which the presence of demarcating stimuli was manipulated. In all conditions, behavior became organized into two-response sequences framed by changes in local reinforcement probability. A tone demarcating the beginning and end of contingent response sequences facilitated the development of functional response units, as in chunking, but the same units developed slowly in the absence of the tone. Complex functional response units developed even though reinforcement contingencies remained constant. These findings demonstrate that models of operant learning must include a mechanism for changing the response unit as a function of reinforcement history. Markov models may seem to be a natural technique for modeling response sequences because of their ability to predict individual responses as a function of reinforcement history; however, no class of Markov chain can incorporate changing response units in their predictions.
Workshop on Data Engineering for Wireless and …, 2001
Studies in African linguistics, 2010
This paper presents the Langi verbal system and the various ways in which tense, aspect and mood ... more This paper presents the Langi verbal system and the various ways in which tense, aspect and mood are encoded. Through a description of the structures and uses of the various forms, it attempts to demonstrate how the different conjugations fit together to form a coherent whole, morphologically and semantically, and how in some cases the system has been influenced by surrounding Cushitic languages.
Computer Systems: Science & Engineering, 2005
This paper presents the Langi verbal system and the various ways in which tense, aspect and mood ... more This paper presents the Langi verbal system and the various ways in which tense, aspect and mood are encoded. Through the description of the structures and uses of the various forms, it attempts to demonstrate how the different conjugations fit together to form a coherent whole, morphologically and semantically, and how in some cases the system has been influenced by surrounding Cushitic languages. RESUME Cet article présente le système verbal du langi et les différents moyens mis en oeuvre pour encoder le temps, l'aspect et le mode. A travers la description des structures et emplois des diverses formes, il tente de démontrer comment les conjugaisons diverses forment un système cohérent, sur les plans morphologiques et sémantiques, et comment, dans certains cas, le système a été influencé par les langues couchitiques environnantes. * I thank the following for their helpful comments on earlier drafts of this paper: Christiane Paulian, Zlatka Guentchéva, Denis Creissels, Dave Odden and an anonymous reviewer at SAL. I am also indebted to Derek Nurse and Maarten Mous for pointing out (as well as providing) various articles of interest for this study. 2 This language is relatively unknown to linguistics: when I began studying it in 1996, the only published work dated from 1916 (by Otto Dempwolff). The data presented here is all first hand, and was gathered during fieldwork I carried out in Tanzania during my doctoral studies, the funding for which was provided by the LACITO-CNRS. Oliver Stegen of SIL has started working on the language recently; so far he has presented a paper on the vowel system at CALL (Leiden) in 2000, and has published a paper on derivation (2002). A monograph on Langi is in press: Dunham (forthcoming).