In the study of emotion recognition, we have compared six typical classifiers by their performances in emotion recognition using physiological signals with daily and personal correction by MANOVA. As the results mentioned above, these clas-sification methods can be very useful to perform emotion recognition by using the physiological signals with daily and personal correction. In particular, we can suc-cessfully correct daily effects using the statistical techniques of MANOVA.
There are still challenges for future studies. For example, we could investigate and determine significant features using feature selection and dimensional reduction methods. In addition, more data collection could be performed in future stud-ies to improve the accuracy. Real-time applications could be further investigated for the prediction of emotional states based on the physiological signals with daily correction. Further adjustments of parameters in classification methods could be investigated. These are interesting topics that we plan to study in the future based on the framework of the current research results.
Regarding the analysis of yeast, five major clusters of gene expression time pro-files were discovered in this study. Four clusters show positive correlations between gene expression profiles in BY and RM strains. The estimated time shifts of expres-sion time profiles in these four clusters are mainly 1 hour after the time that glucose consumption drops. The first cluster shows very interesting pattern of negative
cor-relations between gene expression profiles in BY and RM strains. In this group, the estimated time shift of expression time profiles are mainly 1 hour before the time that glucose consumption drops. These consistent genes show negative correla-tions in two strains are: yor128c, ymr120c, ydr170w-a, yol143c, yor375c-r, ylr346c, yor375c, yhr163w, yor273c, ygr234w. The negative correlations in two strains could be due to the differences of time shifts or the differences in expression shapes in two strains according to the time profiles from microarray data. The experiment data by RT-PCR can be studied to confirm the time profiles of consistent genes in the group of negative correlation of expressions in BY and RM strains in the future. Other models are possible to analyze these microarray data. For instance, time series mod-els with dependent errors, longitudinal modmod-els, modmod-els of functional data analyses and so forth. Besides, network analysis such as Boolean network or Bayesian net-work could be used to investigate the causal relationship of these interesting genes.
These will be of interest to investigate in future studies.
For the study of Time Delay Boolean network, we introduced the Time Delay Boolean network which generalizes the Boolean network model in order to cope dependencies that have time delay relationships. The approach to genetic network inference from gene expression data rely on the assumption that only the expression of a gene is likely to be controlled by a relatively small number (say k) of genes.
Some bounds on the size of data needed for the identification of the Time Delay Boolean networks under constant of indegree are stated. Moreover, the algorithm of the network reconstruction from data with noise are developed.
In practice, there exists differences between real biological systems and Boolean networks. Nodes in a Boolean network take binary values updated synchronously.
In contrast, quantities of gene expression in real cells are continuous and vary with time. Hence, we need to discretize them. The gene expression which is increasing or decreasing with time is also a possible discretization choice.
Work in progress is aimed at evaluating the effectiveness of the described
ap-proach for inferring genetic networks from biological gene expression time series data. Besides, implementation on some other real biological data is also an impor-tant task.
For the implement of the inference algorithm, the most complexity is the com-putation of p-score for each of the k!(n−k)!n! input elements and n output elements, where n is the number of elements and k is the number of indegree. It is an iter-ative algorithm to compute the MLE for the p-scores by E-M procedure and the common practice is setting an upper bound for iterations in numerical implementa-tion. Consequently, this keeps the O(nk+1) complexity for the computation of MLE.
Moreover, the sorting algorithm for the k!(n−k)!n! n data cost O(nk+1log(n)) in time.
Hence, the overall time complexity is O(nk+1log(n)) in this algorithm.
Bibliography
Aha, D. W., Kibler, D., Albert, M. K., 1991. Instanced-based learning algorithms.
Machine Learning 6 (1), 37–66.
Akutsu, T., Kuhara, S., Maruyama, O., Miyano, S., 1998. Identification of gene regulatory networks by strategic gene disruptions and gene overexpression. Proc.
9th ACM-SIAM Symp. Discrete Algorithms, 695–702.
Akutsu, T., Kuhara, S., Maruyama, O., Miyano, S., 2003. Identification of genetic networks by strategic gene disruptions and gene overexpressions under a boolean model. Theoretical computer science 298 (1), 235–251.
Akutsu, T., Miyano, S., 1999. Identification of genetic networks from a small num-ber of gene expression patterns under the boolean network model. Proc. Pacific Symposium on Biocomputing, 17–28.
Ark, W. S., Dryer, D. C., Lu, D. J., 1999. The emotion mouse. International con-ference on Human-computer Interaction, 818–823.
Breazeal, C., Aryananda, L., 2002. Recognition of affective communicative intent in robot directed speech. Autonomous Robots 12, 83–104.
Brem, R. B., Yvert, G., Clinton, R., Kruglyak, L., 2002. Genetic dissection of tran-scriptional regulation in budding yeast. Science 296, 752–755.
Burnham, K. P., Anderson, D. R., 1998. Model selection and inference. Springer, New York.
Busso, C., Deng, Z., Yildirim, S., Bulut, M., Lee, C. M., Kazemzadeh, A., Lee, S., Neumann, U., Narayanan, S., 2004. Analysis of emotion recognition using facial expressions, speech and multinodal information. International Conference on Multimodal Interfaces, 205–211.
Chuang, C. F., Shih, F. Y., 2006. Recognizing facial action units using indepen-dent component analysis and support vector machine. Pattern Recognition 39 (9), 1795–1799.
Cowie, R., Douglas-Cowie, E., Tsapatsoulis, N., Votsis, G., Kollias, S., Fellenz, W., Taylor, J., 2001. Emotion recognition in human-computer interaction. IEEE Signal Processing Magazine 18 (1), 32–80.
Dellaert, F., Polzin, T., Waibel, A., 1996. Recognizing emotion in speech. Interna-tional Conference on Spoken Language Proceedings 3, 1970–1973.
DeRisi, J. L., Iyer, V. R., Brown, P. O., 1997. Exploring the metabolic and genetic control of gene expression on a genomic scale. Science 278 (24), 680–686.
Eisen, M. B., Spellman, P. T., Brown, P. O., Botstein, D., 1998. Cluster analysis and display of genome-wide expression patterns. Proceedings of the National Academy of Sciences of the United States of America 95, 14863–14868.
Fasel, B., Luettin, J., 2003. Automatic facial expression analysis: a survey. Pattern Recognition 36 (1), 259–275.
Friedman, N., Linial, M., Nachman, I., Pe’er, D., 2000. Using bayesian networks to analyze expression data. J. Comp. Biol. 7, 601–620.
Gaffney, S., 2004. Probabilistic curve-aligned clustering and prediction with mixture models. Ph.D. thesis, Department of Computer Science in University of California, Irvine.
Gaffney, S., Robertson, A. W., Smyth, P., Camargo, S. J., Ghil, M., 2007. Probabilis-tic clustering of extratropical cyclones using regression mixture models. Climate Dynamics 29 (4), 423–440.
Gaffney, S., Smyth, P., 2004. Joint probabilistic curve clustering and alignment.
Advances in Neural Information Processing Systems 17, 473–480.
Galindo, C. L., Gadl, A. A., Sha, J., Chopra, A. K., 2004. Microarray an-laysis of aeromonas hydrophila cytotoxic enterotoxin-treated murine primary macrophages. Infection and Immunity 72 (9), 5439–5445.
Gancedo, J. M., 1998. Yeast carbon catabolite repression. Microbiology and Molec-ular Biology Reviews 62 (2), 334–361.
Gasch, A. P., Spellman, P. T., Kao, C. M., Carmel-Harel, O., Eisen, M. B., Storz, G., Botstein, D., Brown, P. O., 2000. Genomic expression programs in the response of yeast cells to environmental changes. Molecular Biology of the Cell 11, 4241–4257.
Heckerman, D., Geiger, D., Chickering, D. M., 1995. Learning bayesian networks:
the combination of knowledge and statistical data. Machine Learning 20, 197–243.
Hu, T., De Silva, L. C., Sengupta, K., 2002. A hybrid approach of nn and hmm for facial emotion classification. Pattern Recognition Letters 23 (11), 1303–1310.
Hunt, E., Marin, J., Stone, P., 1966. Experiments in induction. Academic Press, New York.
Jensen, F., 2001. Bayesian networks and decision graphs. Springer, New York.
Jensen, F. V., 1996. An introduction to bayesian networks. University College Lon-don Press, LonLon-don.
John, G. H., Langley, P., 1995. Estimating continuous distributions in bayesian clas-sifiers. Proceedings of the 11th Conference on Uncertainty in Artificial Intelligence, 338–345.
Kauffman, S. A., 1969. Metabolic stability and epigenesis in randomly constructed genetic nets. Journal of theoretical biology 22 (3), 437–467.
Keerthi, S., Shevade, S., Bhattacharyya, C., Murthy, K., 2001. Improvements to platt’s smo algorithm for svm classifier design. Neural Computation 13 (3), 637–
649.
Kerr, M. K., 2003. Design considerations for efficient and effective microarray stud-ies. Biometrics 59 (4), 822–828.
Kerr, M. K., Churchill, G. A., 2001. Experimental design for gene expression mico-rarrays. Biostatistics 2 (2), 183–201.
Kim, K., Bang, S., Kim, S., 2004. Emotion recognition system using short-term monitoring of physiological signals. Medical and Biological Engineering and Com-puting 42 (3), 419–427.
Lauritzen, S., Spiegelhalt, D., 1988. Local computations with probabilities on graph-ical structures and their application to expert systems. Journal of Royal Statistgraph-ical Society Series B 50, 157–224.
Le Cessie, S., Van Houwelingen, J., 1992. Ridge estimators in logistic regression.
Applied Statistics 41 (1), 191–201.
Li, L. M., Lu, H. S., 2005. Explore biological pathways from noisy array data by directed acyclic boolean networks. Journal of Computational Biology 12 (2), 170–
185.
Littlewort, G., Bartlett, M. S., Fasel, I. R., Chenu, J., Kanda, T., Ishiguro, H., Movellan, J. R., 2004. Towards social robots: Automatic evaluation of human-robot interaction by face detection and expression classification. Advances in Neu-ral Information Processing Systems 16, 1563–1570.
Nasoz, F., Alvarex, K., Lisetti, C. L., Finkelstein, N., 2004. Emotion recognition from physiological signals using wireless sensors for presence technologies. Cogni-tion, Technology and Work 6 (1), 4–14.
Nwe, T., Foo, S., De Silva, L., 2003. Speech emotion recognition using hidden markov models. Speech communication 41, 603–623.
Pearl, J., 1988. Probabilistic reasoning in intelligent systems: networks of plausible inference. Morgan Kaufmann, San Mateo.
Picard, R. W., Vyzas, E., Healy, J., 2001. Toward machine emotional intelligence:
analysis of affective psychological states. IEEE Transactions on Pattern Analysis and Machine Intelligence 23 (10), 1175–1191.
Quinlan, J., 1993. C4.5: Programs for machine learning.
Rani, P., Liu, C., Sarkar, N., Vanman, E., 2006. An empirical study of machine learn-ing techniques for affect recognition in human-robot interaction. Pattern Analysis and Applications 9, 58–69.
Schuller, H.-J., 2003. Transcriptional control of nonfermentative metabolism in the yeast saccharomyces cerevisiae. Current Genetics 43 (3), 139–160.
Spellman, P. T., Sherlock, G., Zhang, M. Q., Iyer, V. R., Anders, K., Eisen, M. B., Brown, P. O., Botstein, D., Futcher, B., 1998. Comprehensive identification of cell cycle-regulated genes of the yeast saccharomyces cerevisiae by microarray hybridization. Molecular Biology of Cell 9, 3273–3297.
Vapnik, V. N., 1998. Statistical learning theory. Wiley, New York.
Yacoob, Y., Davis, L. S., 1996. Recognizing human facial expression from long image sequences using optical flow. IEEE Transactions on Pattern Analysis and Machine Intelligence 18 (6), 636–642.
Zhou, C., Lin, X., 2005. Facial expressional image synthesis controlled by emotional parameters. Pattern Recognition Letters 26 (16), 2611–2627.