9 results on '"Data Reduction"'
Search Results
2. TÜRKİYE’DE ENFLASYON ORANININ TEMEL BİLEŞENLİ LP NORM YÖNTEMİ İLE TAHMİNİ
- Author
-
Öznur İşçi̇ and Atillla Göktaş
- Subjects
çıkarım süreci ,enflasyon ,enflasyonu etkileyen faktörler ,i̇leri dönük kestirim ,lp-norm ,rank sorunu ,temel bileşenler ,veri i̇ndirgeme ,zaman serisi ,data reduction ,forecasting ,inflation ,principal components ,rank problem ,statistical inference ,the factors of inflation rate ,time series ,Business ,HF5001-6182 - Abstract
Enflasyon belli bir süre zarfında ekonomideki mal ve hizmetlerin genel fiyat düzeylerinin yükselmesidir. Bir ülkedeki yüksek düzeyli ve sürekli enflasyon gerek toplumu gerekse de ülkenin ekonomisini negatif etkilemektedir. Bu da o ülkede uzun dönem için doğru kararlar alınmasına engel olur. Enflasyon oranının yüksek düzeylerde olması sebebiyle, sosyal yapı ve ülkenin rekabeti negatif yönde etkilenecektir. Bu gerçekler doğrultusunda, enflasyon oranının belirlenmesi ve gerekli önlemlerin alınması kaçınılmaz bir gerekliliktir. Enflasyon oranını tahminlemede farklı yöntemler kullanılmaktadır. Normal dağılım varsayımları istatistik literatüründe genellikle kullanılmaktadır. Fakat enflasyon oranının tahminlemesi üzerine çalışan bir istatistikçi varsayımların bozulması durumunu nasıl düzenleyebileceğini bilemeyebilir. Varsayımlar sağlanmadığından dolayı, istatistiksel çıkarım geçekleştirilemez. Bu gerçekler doğrultusunda, istatistiksel çıkarım süreçlerinin gerçekleştirilmesi için genel hata dağılışı ya da Lp-Norm olarak bilinen üstel kuvvet dağılışı kullanılmıştır. Üstel kuvvet dağılışında, varsayımların sağlanması için optimal p değerinin seçilme zorunluluğu vardır. Bu çalışmanın diğer bir amacı enflasyonu etkileyen değişkenlerdeki çoklu doğrusal bağlantı sorunun ortadan kaldırılması ve bunun yanında temel bileşenler analizi kullanılarak bağımsız değişkenlerdeki bilgi yapısının çoğunu açıklayan bu faktörlerden yeni ilişkisiz değişkenler elde etmektir. Açıklayıcı değişken olan bu yeni değişkenler enflasyon regresyon modelinin oluşturulmasında kullanılmaktadır. Üstel kuvvet dağılışındaki tahminlenmiş olan p değeri model belirlemede kullanılmaktadır. Bu p değeri ile belirlenen model Türkiye’de enflasyon oranını öngörmede kullanılmaktadır.
- Published
- 2011
3. Dik eşleştirme arayış yöntemi ile hibrit veri sıkıştırma ve optiksel kriptografi.
- Author
-
Atar, Ertan, Ersoy, Okan K., and Özyılmaz, Lale
- Subjects
- *
DATA compression , *ORTHOGONAL matching pursuit , *CRYPTOGRAPHY , *COMPRESSED sensing , *DATA reduction , *DATA encryption - Abstract
Compressive Sensing (CS), which makes it possible to reduce the amount of data and thereby to greatly simplify the sensor system has become a very important research area. In this method, data is compressed before measurements whereas data is first measured and then compressed in the current technology. This approach leads to reducing the number of sensors. In this study, simultaneous compressive sensing and hybrid optical cryptography is developed as a new approach to handle two important problems in the field of communications, namely, effective and efficient signal processing and secure transmission of information. In this work, CS is achieved with the orthogonal matching pursuit (OMP) algorithm. Then, the CS-OMP algorithm is combined with the double random phase encyrption (DRPE) method to achieve both compression and encyrption of data. In order to achieve high security, the keys used in CS and DRPE are transmitted to the receiver by an asymmetric cryptography method. Thus, the overall cryptographic system is a hybrid optical system (both symmetric and asymmetric) since DRPE is a symmetric optical encyrption method. [ABSTRACT FROM AUTHOR]
- Published
- 2017
- Full Text
- View/download PDF
4. Buhar sıkıştırmalı soğutma çevrimi şarj dağ#x0131;lımının deneysel incelenmesi.
- Author
-
Özyurt, Bekir and Nilüfer^Egrıcan, A.
- Subjects
- *
DYNAMO (Computer program language) , *EVAPORATORS , *CAPACITORS , *STEAM condensers , *REFRIGERATION & refrigerating machinery , *REFRIGERANTS , *DATA reduction , *COMPUTER software , *HEAT exchangers - Abstract
Vapor compression refrigeration cycle is widely used in industrial and residential refrigeration systems. The system performance is seriously affected by refrigerant selection and amount used as well as choice of cycle components like compressor, evaporator, and condenser. One major milestone for the refrigeration industry is discovery of CFC's at 1930's. CFC's make excellent choice for vapor compression refrigeration with their high molecular weight and non-flammability properties. At 1987 the use of CFC's are limited because of their effects on Ozone layer by Montreal Protocol. In 1990 R134a is launched as a HFC class refrigerant, but the HFC's still have a high global warming potential value beside their suitable ozone depletion potential values. The following years HC's and specially R600a are became popular for small refrigeration systems, because only negative property of R600a is flammability is not a big concern in small amounts. In this study, for better understanding of vapor compression refrigeration system working conditions, the charge distribution at the system components and effects of different parameters to this distribution is investigated with R600a refrigerant. The charge distribution measurement methods expect complicated and expensive ones depend on weighting methods. The separate and weight method usually needs more time and workforce. Two experimental setups are prepared for this purpose; first one is extended version of balance system in literature with replacing balance with load cells that have a higher precision. This system gives rapid results for the measurement but needs a data reduction process for frost formation and waves formed by hanging system. Second setup is based on principle of expanding the refrigerant, closed in components by quick closing valves at a specified moment of the cycle. The refrigerant amount can be easily calculated from pressure and temperature values for the expanded gas that have superheated condition. A computer program developed in the study controls the pressure stabilization, reaching steady conditions, closing of valves at the desired data point and expanding of component charges to the tank one by one. The measurement of one data point takes nearly 8-10 hours, and measurement of an on-off cycle with lots of data points take more than a week. But the data obtained are more precise than the other system and includes results about dryer and compressor gas volume. The refrigerant distribution is measured under steady state and cyclic working conditions, and effect of parameters like total charge value, use of variable capacity compressor and different cycle times with these setups. The experiments for different total charge values show that the maximum change is in evaporator. … [ABSTRACT FROM AUTHOR]
- Published
- 2010
5. Determination of inflation rate in Turkey using lp norm method with principal components
- Author
-
İşçi, Öznur, Göktaş, Atilla, and TR199306
- Subjects
İleri dönük kestirim ,Principal components ,Time series ,Data reduction ,Lp-norm ,Rank problem ,The factors of inflation rate ,Enflasyonu etkileyen faktörler ,Inflation ,Çıkarım süreci ,Veri indirgeme ,Temel bileşenler ,Enflasyon ,Rank sorunu ,Zaman serisi ,Statistical inference ,Forecasting - Abstract
Enflasyon belli bir süre zarfında ekonomideki mal ve hizmetlerin genel fiyat düzeylerinin yükselmesidir. Bir ülkedeki yüksek düzeyli ve sürekli enflasyon gerek toplumu gerekse de ülkenin ekonomisini negatif etkilemektedir. Bu da o ülkede uzun dönem için doğru kararlar alınmasına engel olur. Enflasyon oranının yüksek düzeylerde olması sebebiyle, sosyal yapı ve ülkenin rekabeti negatif yönde etkilenecektir. Bu gerçekler doğrultusunda, enflasyon oranının belirlenmesi ve gerekli önlemlerin alınması kaçınılmaz bir gerekliliktir. Enflasyon oranını tahminlemede farklı yöntemler kullanılmaktadır. Normal dağılım varsayımları istatistik literatüründe genellikle kullanılmaktadır. Fakat enflasyon oranının tahminlemesi üzerine çalışan bir istatistikçi varsayımların bozulması durumunu nasıl düzenleyebileceğini bilemeyebilir. Varsayımlar sağlanmadığından dolayı, istatistiksel çıkarım geçekleştirilemez. Bu gerçekler doğrultusunda, istatistiksel çıkarım süreçlerinin gerçekleştirilmesi için genel hata dağılışı ya da Lp-Norm olarak bilinen üstel kuvvet dağılışı kullanılmıştır. Üstel kuvvet dağılışında, varsayımların sağlanması için optimal p değerinin seçilme zorunluluğu vardır. Bu çalışmanın diğer bir amacı enflasyonu etkileyen değişkenlerdeki çoklu doğrusal bağlantı sorunun ortadan kaldırılması ve bunun yanında temel bileşenler analizi kullanılarak bağımsız değişkenlerdeki bilgi yapısının çoğunu açıklayan bu faktörlerden yeni ilişkisiz değişkenler elde etmektir. Açıklayıcı değişken olan bu yeni değişkenler enflasyon regresyon modelinin oluşturulmasında kullanılmaktadır. Üstel kuvvet dağılışındaki tahminlenmiş olan p değeri model belirlemede kullanılmaktadır. Bu p değeri ile belirlenen model Türkiye’de enflasyon oranını öngörmede kullanılmaktadır., In economics, inflation is a rise in general level of prices of goods and services in an economy over a period of time. High leveled continuous inflation in one country has a negative influence on negatively both the community and economy of the country. A country with high leveled inflation has an obstacle on the right decision for long period. Hence, social structure and competition of the country are affected in negative direction. In the light of these facts, the determination of inflation rate and precaution must be necessarily taken. Several methods are used to estimate the inflation rate. The assumptions of normal distribution is commonly used in statistical literature but a statistician studying the estimation of inflation rate do not know how the violation those assumptions can be arranged. On account of the fact that these assumptions are not guaranteed, the statistical inference is not carried out. As a result of these facts, we use the exponential power distribution known as general error distribution or Lp-Norm to carry out the statistical inference procedures. In the exponential power distribution, we need to choose the optimal value of p to guarantee the assumptions. The aim of this study is to remove the collinearity among regressors that affect the inflation and besides using the principal components we obtain independent factors that are explaining most of the information of the structure of the regressors. The new regressors are used in constructing of regression model for inflation. Estimated p value in exponential power distribution is used to determine a regression model to forecast the inflation rate in Turkey.
- Published
- 2011
6. E-museum: web-based tour and information system for museums
- Author
-
Baştanlar, Y., Altıngövde, İsmail Şenol, Aksay, A., Alav, O., Çavuş, Özge, Yardımcı, Y., Ulusoy, Ozgur, Güdükbay, Uğur, Çetin, A. Enis, Akar, G. B., Aksoy, Selim, Çetin, A. Enis, and Güdükbay, Uğur
- Subjects
World Wide Web ,Data reduction ,Museums ,Retrieval modules ,Visual information ,Information systems ,Visual contents ,Web browsers ,GeneralLiterature_MISCELLANEOUS ,Virtual reality ,Artworks - Abstract
Date of Conference: 17-19 April 2006 A web-based system - consisting of data entrance, access and retrieval modules - is constructed for museums. Internet users that visit the e-museum, are able to view the written and visual information belonging to the artworks in the museum, are able to follow the virtual tour prepared for the different sections of the museum, are able to browse the artworks according to certain properties, are able to search the artworks having the similar visual content with the viewed artwork. © 2006 IEEE.
- Published
- 2006
7. Improvement of face detection algorithms for news videos
- Author
-
Ikizler, Nazlı and Duygulu, Pınar
- Subjects
Video signal processing ,Face detection algorithms ,News videos ,Data reduction ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION ,Schneiderman-Kanade face detection ,Face recognition ,Algorithms ,Image analysis ,Spurious signal noise - Abstract
Date of Conference: 16-18 May 2005 People are the most important subjects in news videos and for proper retrieval of person images, face detection is a very crucial step. However, face detection and recognition in news videos is a very challenging task due to the huge irregularities and high noise level in the data. This study presents a method that combines skin detection and Schneiderman-Kanade face detection, for improving the face detection performance in news videos for a better retrieval. This method has been tested on TRECVID 2003 dataset and the results are very promising. © 2005 IEEE.
- Published
- 2005
8. Modeling rural road transport demand based on genetic algorithm
- Author
-
Ceylan, Halim and Haldenbilen, Soner
- Subjects
Stochastic search process ,Mathematical models ,Motor transportation ,Data reduction ,Economic and social effects ,Strategic planning ,Genetic Algorithm Travel Demand (GATT) ,Random processes ,Genetic algorithms ,Gross National Product (GNP) ,Transport demand planning ,Rural roads - Abstract
The paper describes the use of stochastic search process that is the basis of Genetic Algorithms (GAs) in developing transport demand in rural roads of Turkey. Travel demand, vehicle movements and demand of goods transport are estimated based on the socioeconomic indicators that are the Gross National Product (GNP), population and number of vehicles. Various forms of the Genetic Algorithm Travel Demand (GATT) models are developed. Weighting parameters of the GATT models are estimated using the historical data. Available data is partly used for estimating weighting parameters of the GATT and partly for testing the models. GATT models are validated with observed data, and then future estimation of travel demand is projected until 2025. Results are compared with European Union (EU) countries and suggestions are made in the long term in terms of transport demand planning.
- Published
- 2005
9. Birincil dizi veri temelli protein hücre içi yer belirleme tahmini
- Author
-
Mert Özarar, Volkan Atalay, and Rengul Cetin Atalay
- Subjects
Self-organizing map ,Statistical methods ,Computer science ,Cells ,Computational biology ,Signal encoding ,Bioinformatics ,medicine.disease_cause ,Matrix algebra ,Substitution matrix ,medicine ,Amino acid content ,Cluster analysis ,Sequence (medicine) ,Self organizing maps ,chemistry.chemical_classification ,Mutation ,Problem solving ,Data reduction ,business.industry ,Proteins ,Pattern recognition ,Perceptron ,Subcellular localization ,Amino acid ,Order (biology) ,chemistry ,Four class problem ,Encoding scheme ,Amino acids ,Artificial intelligence ,Primary sequence ,business ,Protein subcellular localization (P2SL) ,Classifier (UML) ,Function (biology) - Abstract
Date of Conference: 28-30 April 2004 Conference Name: IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004 Subcellular localization is crucial for determining the functions of proteins. A system called prediction of protein subcellular localization (P2SL) that predicts the subcellular localization of proteins in eukaryotic organisms based on the amino acid content of primary sequences using amino acid order is designed. The approach for prediction is to find the most frequent motifs for each protein in a given class based on clustering via self organizing maps and then to use these most frequent motifs as features for classification by the help of multi layer perceptrons. This approach allows a classification independent of the length of the sequence. In addition to these, the use of a new encoding scheme is described for the amino acids that conserves biological function based on point of accepted mutations (PAM) substitution matrix. The statistical test results of the system is presented on a four class problem. P2SL achieves slightly higher prediction accuracy than the similar studies.
- Published
- 2004
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.