Bulgulayıcı veri analizi

dc.contributor.advisor Şenesen, Ümit
dc.contributor.author Demirbilek, Emel
dc.contributor.authorID 19352
dc.contributor.department Matematik Mühendisliği tr_TR
dc.date.accessioned 2023-03-03T13:03:56Z
dc.date.available 2023-03-03T13:03:56Z
dc.date.issued 1991
dc.description Tez (Yüksek Lisans) -- İstanbul Teknik Üniversitesi, Fen Bilimleri Enstitüsü, 1991 tr_TR
dc.description.abstract Yapılan her araştırma sonucunda, o konuyu açıklayıcı sayılara ulaşılır. Bu sayılar, istatistik! analiz yöntemleri kullanılarak işlenir ve sayılardan anlamlı sonuçlara varılır. Böyle "bir çalışma zamanı, "belirli "bir düzeyde matematik "bilgisini, formülleri ve hesaplamaları gerektirir. Oysa,bazı durumlarda kısa süreli "bir çalışma ile yorum yapılabilecek sonuçlara ulaşmak istenebilir. Tukey tarafından geliştirilen bulgulayıcı veri analizi yöntemlerinin "bir kısmı kullanılarak hazırlanan hu programlar, fazla hassas olmayan gözlem değerleri ile yorum yapılabilir defterlere ulaştırır. Programlar COBOL programlama diliyle yazılmıştır. Tüm programlar PC ortamında ve DOS işletim sistemi altında çalışır. İlk olarak veri gruplarındaki eleman sayısı, yuvarlama basamağı gibi ortak parametreler girişi yapılır. Gruplara ait gözlem değerleri parametrelere göre işlenerek, dosyalara kaydedilir. İstenilen sayıda veri grubu girişi yapılabilir. İşlem bölümünde seçilen iki veri grubu kullanılarak, sayısal esas değerler. hesaplanıp, del-ve-yaprak, kutu-ve -nokta, standart kutu-ve-nokta çizimleri yapılır. Amaç, tüm bu hesaplama ve çizimler yardımı ile kullanıcının rahat bir yorum yapmasına yardımcı olmaktır. tr_TR
dc.description.abstract In these days, the different kinds of topics are searched and the result of researches, are expressed by numbers. The science of statistics make the results understandable and clear, but it is not easy to analyze data. Firstly organizing numbers are difficult, it requires mathematical knowledge and calculations take time. Using exploratory data analysis techniques is easier and quicker and more enjoyable than ythe other techniques. Calculations are simple, and the measurements used are not so sensitive. Exploratory methods make it easy to see into the data. Using the computer for these calculations, cut down the processing time and make the results healtier. First of all the data are given to the computer and then processed by programmes. Finally the results are given to the operator by using sohedules and graphics. This program package is prepared to be used for exploratory data analysis. These programs are written in COROL (COmmon Business Oriented Language) program language. All programmes work on Personal Computer environment and under the DOS operating system. Printed outputs are written by the eigthy characters printer file. The package censists of twelve programs - MENU - CAGIR - XEZ1 - XEZ2 - XEZ3 - XEZ31 - v - - XEZ4 - XEZ7 - XEZ17 - XEZ8 - XEZ10 - XEZ11 The programme package has basicly four steps : - To give conmon parameters of the data groups, - to give observations of the data groups, - to process data, - to draw tables and graphies by using results. Firstly the programme nemed MENU which maneges the package works. At the first attempt the parameters and then at least two data groups are given. Unless these have heen done, calculation programmes can not work. First of all parameters are given by using the programme named XEZ1. Each data groups has the same number of observations which can not be more than a hundered. To simplify numbers, they can be rounded by truncating method or original rates can be kept. Secondly, the observations of the data group are entered by using the programme named XEZ2. After all the data put in, they are ordered from minimum to maximum. These numbers are rounded and separated for stem and leaf one by one. They are written on the disk file which is named by user at the beginning of the programme. The numbers of data groups are unlimited but comparison can be made by only two batches of data at a time. Thirdly, the names of two chossen batches of data are given for calculations and comparison by using the programmes named CAGIR. CAGIR calls XEZ3, XEZ31, XEZ4, XEZ7, XEZ17, XEZ8, XEZ10, XEZ11 step by step. XEZ3 displays records of each data files. - vi - xEZ31 works for the same purpose of XEZ3 but prints the list. XEZ4 computes the numerical summaries. These numerical summary values are : - Median, - Quartiles, - Range, - Midspread, - Step, - Outliers, - Far outliers, - Maximum observation value, - Minimum observation value. They are displayed and written on the parameter's file. XEZ7 draws stem-and-leaf graph on the screen to compare two batches of data. Under the graphs numerical summaries are displayed. XEZ17 works for the same purpose of XEZ7 but prints the graphs and numerical summaries. XEZ8 which draws box-and-dot plot, runs once for each batches. XEZ10 displays standerdized box-and-dot plot. Under the plot standardized numerical summaries are displayed. XEZ11 works for the same purpose of XEZ10 but prints the plot and summaries. After running of XEZ11, it returns to MENU. There are three choices : - To put new data groups in, - To choose new batches for the third step, - To exit programme. - vii - Two kinds of disk file are used in programmes: Sequential and indexed files. Parameter file named PARAM.DAT consists of parameter values and numerical summaries of two batches. File's organization and access mode are sequential. Variables of file are : - The title of the subject, - The title of the first batch, - The title of the second batch, - The title which is the kind of subject, - The count of observations, - Rounding scale, - Numerical summaries which belong to batches : - Median, - Upper quartile, - Lower quartile, - Upper outliers, - Lower outliers, - Far upper outliers, - Far lower outliers. data file's pame is the same as the title of the batches. File's organization is indexed and access made is dynamic. Variables of file are : - The record number, - The observation value, - The explanation of observation, - The stem of observation, - The leaf of observation. - vm stem-and-teaf standardized box-and-dot Flow Chard of the Programmes en_US
dc.description.degree Yüksek Lisans tr_TR
dc.identifier.uri http://hdl.handle.net/11527/22358
dc.language.iso tr
dc.publisher Fen Bilimleri Enstitüsü tr_TR
dc.rights Kurumsal arşive yüklenen tüm eserler telif hakkı ile korunmaktadır. Bunlar, bu kaynak üzerinden herhangi bir amaçla görüntülenebilir, ancak yazılı izin alınmadan herhangi bir biçimde yeniden oluşturulması veya dağıtılması yasaklanmıştır. tr_TR
dc.rights All works uploaded to the institutional repository are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. en_US
dc.subject Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol tr_TR
dc.subject Veri analizi tr_TR
dc.subject Computer Engineering and Computer Science and Control en_US
dc.subject Data analysis en_US
dc.title Bulgulayıcı veri analizi
dc.title.alternative Exploratory data analysis
dc.type Tez tr_TR
Dosyalar
Orijinal seri
Şimdi gösteriliyor 1 - 1 / 1
thumbnail.default.alt
Ad:
19352.pdf
Boyut:
3.13 MB
Format:
Adobe Portable Document Format
Açıklama
Lisanslı seri
Şimdi gösteriliyor 1 - 1 / 1
thumbnail.default.placeholder
Ad:
license.txt
Boyut:
3.16 KB
Format:
Plain Text
Açıklama