The choice of the topology of neural networks and their use for the classification of small texts

O.S. Smirnova, V.V. Shishkov

Abstract


In this article, we describe the classification of texts as the approaches used to gain knowledge from unstructured data. This approach includes determining messages tone. We describe the choice of the topology for neural networks and the methodology for classifiers creation. Our paper presents testing results for two created classifiers

Full Text:

PDF (Russian)

References


Richard Alleyne. Welcome to the information age. Jel. resurs: http://www.telegraph.co.uk/news/science/science-news/8316534/Welcome-to-the-information-age-174-newspapers-a-day.html (data obrashhenija: 17.05.2016 g.).

Stratonovich R.L. Teorija informacii. — M.: Sovetskoe radio, 1975.

Moore, Gordon E. No Exponential is Forever: But «Forever» Can Be Delayed!. International Solid-State Circuits Conference (ISSCC) 2003 / SESSION 1 / PLENARY / 1.1.

Keen, P. G. W. (1978). Decision support systems: an organizational perspective. Reading, Mass., Addison-Wesley Pub. Co. ISBN 0-201-03667-3

Wikipedia. Statistical classification. Jel. resurs: https://en.wikipedia.org/wiki/Statistical_classification (data obrashhenija: 17.05.2016 g.).

D.Bazhenov. Naivnyj bajesovskij klassifikator. Jel. resurs: http://bazhenov.me/blog/2012/06/11/naive-bayes.html (data obrashhenija: 23.05.2016 g.).

Vincent A. Akpan Adaptive predictive control using recurrent neural network identification. Jel. resurs: http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5164515&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D5164515 (data obrashhenija: 17.05.2016 g.).

Wikipedia. Recurrent neural network. Jel. resurs: https://en.wikipedia.org/wiki/Recurrent_neural_network (data obrashhenija: 17.05.2016 g.).

Richard M. Zur, Yulei Jiang, Lorenzo L. Pesce, and Karen Drukker. Noise injection for training artificial neural networks: A comparison with weight decay and early stopping. Jel. resurs: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2771718/ (data obrashhenija: 23.05.2016 g.).

Wikipedia. Nejronnye seti / Vybor topologii seti. Jel. resurs: https://ru.wikipedia.org/wiki/Iskusstvennaja_nejronnaja_set'#.D0.92.D1.8B.D0.B1.D0.BE.D1.80_.D1.82.D0.BE.D0.BF.D0.BE.D0.BB.D0.BE.D0.B3.D0.B8.D0.B8_.D1.81.D0.B5.D1.82.D0.B8 (data obrashhenija: 25.05.2016 g.).

Habrahabr. Obuchenie OpenCV kaskada Haara. Jel. resurs: https://habrahabr.ru/post/208092/ (data obrashhenija: 25.05.2016 g.).

NVIDIA Parallel'nye vychislenija CUDA. Jel. resurs: http://www.nvidia.ru/object/cuda-parallel-computing-ru.html (data obrashhenija: 27.05.2016 g.).

Wikipedia. Nejronnye seti / Izvestnye tipy setej. Jel. resurs: https://ru.wikipedia.org/wiki/Iskusstvennaja_nejronnaja_set'#.D0.98.D0.B7.D0.B2.D0.B5.D1.81.D1.82.D0.BD.D1.8B.D0.B5_.D1.82.D0.B8.D0.BF.D1.8B_.D1.81.D0.B5.D1.82.D0.B5.D0.B9 (data obrashhenija:27.05.2016 g.).

Wikipedia. Svertochnaja nejronnaja set'. Jel. resurs: https://ru.wikipedia.org/wiki/Svertochnaja_nejronnaja_set' (data obrashhenija: 29.05.2016 g.).

Wikipedia. Nejronnaja set' Jelmana. Jel. resurs: https://ru.wikipedia.org/wiki/Nejronnaja_set'_Jelmana (data obrashhenija: 27.05.2016 g.).

Russell C. Eberhart, Roy W. Dobbins – Neural Network PC Tools: A Practical Guide – Academic Press, 28 Jun. 2014 – pp. 90 – 134.

Wikipedia. BFGS. Jel. resurs: https://en.wikipedia.org/wiki/Broyden–Fletcher–Goldfarb–Shanno_algorithm (data obrashhenija: 29.05.2016 g.).

Wikipedia. CG. Jel. resurs: https://en.wikipedia.org/wiki/Conjugate_gradient_method (data obrashhenija: 27.05.2016 g.).

Portal tartarus.org Russian stemming algorithm. Jel resurs: http://snowball.tartarus.org/algorithms/russian/stemmer.html (data obrashhenija: 29.05.2016 g.).

Portal keras.io Dokumentacija Keras. Jel. resurs: http://keras.io/getting-started/faq/#why-is-the-training-loss-much-higher-than-the-testing-loss (data obrashhenija: 27.05.2016 g.).

Ju.V.Rubcova. Postroenie korpusa tekstov dlja nastrojki tonovogo klassifikatora. Programmnye produkty i sistemy, 2015, #1(109), – S.72-78.

Github, hosting proektov Pystemmer Jel. resurs: https://github.com/snowballstem/pystemmer (data obrashhenija: 25.05.2016 g.).

Portal Algorithmist. Stop simvoly russkogo jazyka. Jel. resurs: http://www.algorithmist.ru/2010/12/stop-symbols-in-russian.html (data obrashhenija: 27.05.2016 g.).

Portal deeplearning.net. Dokumentacija. Theano Jel. resurs: http://deeplearning.net/software/theano/ (data obrashhenija: 27.05.2016 g.).


Refbacks

  • There are currently no refbacks.


Abava  Кибербезопасность IT Congress 2024

ISSN: 2307-8162