Работа с базами данных
<<  Data Mining: Exploring Data Правила и процедуры УНТЦ  >>
420,000 Million is Lot
420,000 Million is Lot
420,000 Million is Lot
420,000 Million is Lot
420,000 Million is Lot
420,000 Million is Lot
2008 and 2009 Non-CAN Vehicles
2008 and 2009 Non-CAN Vehicles
Support for A/C and Heated Catalyst Monitors
Support for A/C and Heated Catalyst Monitors
Support for A/C and Heated Catalyst Monitors
Support for A/C and Heated Catalyst Monitors
Картинки из презентации «What Can Your Data Tell You» к уроку информатики на тему «Работа с базами данных»

Автор: Michael St. Denis. Чтобы познакомиться с картинкой полного размера, нажмите на её эскиз. Чтобы можно было использовать все картинки для урока информатики, скачайте бесплатно презентацию «What Can Your Data Tell You.pptx» со всеми картинками в zip-архиве размером 442 КБ.

What Can Your Data Tell You

содержание презентации «What Can Your Data Tell You.pptx»
Сл Текст Сл Текст
1What Can Your Data Tell You? 8Manufacturer Identifier. World
Interesting Observations! Robert Budd, Manufacturer Identifier. World
Michael St. Denis, D.Env., Joe Roeschen - Manufacturer Identifier. Vehicle
Revecorp Inc. I/M Solutions May 5-8, 2013 Attributes. Vehicle Attributes. Vehicle
Schaumberg, Illinois. Attributes. Vehicle Attributes. Vehicle
2Background. Vehicle data has become Attributes. Check Digit. Model Year. Plant
critical to accurate testing and fraud Code. Sequential Number. Sequential
detection There are many general rules Number. Sequential Number. Sequential
about OBDII vehicles we all use “Trust but Number. Sequential Number. Sequential
Verify” – we wanted to confirm the rules Number. 2. G. 4. W. S. 5. 2. M. W. 1. 1.
with I/M field data observations We found 2. 3. 4. 5. 6. 7. 8. 10. 11. World
some very interesting and important things Manufacturer Identifier. World
Some vehicles in the fleet that were Manufacturer Identifier. World
surprising (some already known, some Manufacturer Identifier. World
possibly not) Examples important to I/M Manufacturer Identifier. World
program operations will be given I/M data Manufacturer Identifier. World
“truths” must be understood and applied by Manufacturer Identifier. Vehicle
VIN in order to be accurate. Revecorp Inc. Attributes. Vehicle Attributes. Vehicle
3Background. Needed a large data source Attributes. Vehicle Attributes. Vehicle
to develop rigorous statistical confidence Attributes. Vehicle Attributes. Vehicle
in the results Revecorp has been Attributes. Vehicle Attributes. Vehicle
collecting I/M Data for a very long time – Attributes. Vehicle Attributes. Model
and “normalized” them so they could Year. Model Year. Plant Code. Plant Code.
analysis could be performed across Revecorp Inc.
programs A lot of data, that is very 9VIN Stem. year. make. division. model.
powerful and you can learn a lot from if base. style. Veh typ. 1998. GM. BUICK.
you look carefully How much data? CENTURY CUSTOM. CENTURY. SEDAN 4 DOOR. P.
Approaching half a billion records. How Body type. drive. cylinders. liters.
big is that??? There are about 150 people block. 4D. F. 6. 3.1. V. Fuel type. GVWR
at the conference, and we want each of you lbs. CURB lbs. LVW lbs. ALVW lbs. Mobile6
to take some of our data home. So we are class. G. 4464. 3255. 3555. 3860. LDGV.
going to print it out for you…. Revecorp Revecorp Inc.
Inc. 10Electronic VIN. All OBDII vehicles in
4420,000 Million is Lot. Record Count. the US are supposed to provide electronic
420,000,000. records. Sheets Per Ream. VIN starting in 2005 Some manufacturers
500. sheets. Reams per case. 10. reams. started as early as 2000 For model years
Cases per pallet. 40. cases. Inches per 2000 to 2004, this can be used to indicate
ream. 2.25. inches. Attendees. 150. fraud (the vehicle should support e-vin,
people. Records per Attendee. 2,800,000. but nothing is received from the vehicle)
records. Reams per Attendee. 5,600. reams. Examples 2001 Ford Focus will return a
Cases per Attendee. 560. cases. Pallets valid eVIN (valid result will pass
per Attendee. 14. pallets. Height of stack checksum calc) 2001 Ford Taurus will
per Attendee. 1,050. feet. Revecorp Inc. return invalid eVIN (junk response won’t
5420,000 Million is Lot. Sears (Willis) pass checksum) 2001 Ford Mustang will
Tower is 1,260 feet tall. Revecorp Inc. return null eVIN (blank). Revecorp Inc.
6Common “General Rules”. If we know the 11Now that’s odd… Revecorp Inc.
rules vehicle OBDII data should follow, 122008 and 2009 Non-CAN Vehicles. There
when the collected data does not match are some 2008 and one 2009 model year
these rules, we suspect there is fraud. vehicle which are not CAN (55 VIN stems
There are several easy examples: Cars total). Revecorp Inc.
started supporting e-VIN electronically in 13Support for A/C and Heated Catalyst
2003 All vehicles use CAN communication Monitors. Revecorp Inc.
protocol starting in 2008 No vehicles 14Results and Conclusions. With the use
support the air conditioning monitor (in of a large data set, good guidance on
fact an OBDII simulator commonly used by testing and fraud prevention can be
I/M programs for auditing is set to provided Knowing these is important to
support the A/C monitor to indicate it is avoiding mistakes in OBDII testing Common
NOT a real vehicle) No vehicles support standards for storing OBDII data would be
the heated catalyst monitor EPAs readiness helpful - The Remote OBDII guidance
exclusion criteria should be specific to document could be used as a start Please
the impacted vehicles. Revecorp Inc. collect information on the specific type
7How Was the Data Analyzed. There are of communication protocol successfully
many different naming conventions used in used (CAN 11/29 bit, KWP fast or slow
the automotive industry –EPA/Sierra, ACES, initialization, etc.) We merged the OBDII
Polk, VinPower, NCIB, Federal It is failure data with recall data and found
difficult to merge data by name RAV4, many vehicles have had recalls for many
RAV/4, RAV 4, RAV-4, …… OEM conventions years, and clearly that information is not
used from their VIN decoding submissions being communicated to motorists – but it
as per 49 CFR Part 565 We use VIN “Stems” would help. Revecorp Inc.
to look at data for a common vehicle Using 15Can We Have Your Data? We would like
VIN stem matching versus Year/Make/Model to assist in answering questions about
gives perfect matches and great data OBDII data observed in the field Many
resolution. Revecorp Inc. programs have been very generous with
8VIN Stem (squish vins). 1. 2. 3. 4. 5. their data, we would like to work with
6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. each program to obtain your I/M data We
17. World Manufacturer Identifier. World will take it in any electronic form We
Manufacturer Identifier. World will only use the data in aggregate not
Manufacturer Identifier. Vehicle disclose the source of the data, but will
Attributes. Vehicle Attributes. Vehicle share the results with you if we find it
Attributes. Vehicle Attributes. Vehicle anomalous (i.e. one piece of equipment is
Attributes. Check Digit. Model Year. Plant providing incorrect data, your program is
Code. Sequential Number. Sequential showing different data than others, etc.).
Number. Sequential Number. Sequential Revecorp Inc.
Number. Sequential Number. Sequential 16Questions? Contact: Robert Budd 5732
Number. 2. G. 4. W. S. 5. 2. M. 8. W. 1. Lonetree Blvd Rocklin, CA 95765 (916)
5. 3. 2. 0. 0. 1. 1. 2. 3. 4. 5. 6. 7. 8. 786-1006 x1004 robert.budd@revecorp.com.
9. 10. 11. 12. 13. 14. 15. 16. 17. World Revecorp Inc.
What Can Your Data Tell You.pptx
cсылка на страницу

What Can Your Data Tell You

другие презентации на тему «What Can Your Data Tell You»

«English for you» - Все слова и выражения озвучены носителями языка. Ты убедишься насколько интересным и захватывающим может быть обучение языку. Слова и выражения по темам. Узнать насколько хорошо ты усвоил материал тебе помогут: Может ли компьютер заменить учителя? Викторины Задания Игры Игровые упражнения Ты можешь выбрать уровень сложности.

«Data Mining» - Оценка параметра k методом кросс-проверки. Примеры использования и реализации. Понятие Data Mining. Data Mining. Стадии Data Mining. История и основные понятия. Высокая зависимость результатов классификации от выбранной метрики. Преимущества метода. Задачи Data Mining. Мультидисциплинарность. Анализ связей (корреляционный и регрессионный анализ, факторный анализ, дисперсионный анализ).

«Триггеры баз данных» - Создание замещающих триггеров. Пример замещающего триггера. Имя отдела. Триггер замещения. Понятие триггера. Представления, которые содержат соединения. Привилегии для создания триггера. Создание замещающих триггеров баз данных. Модифицируемые и немодифицируемые представления. Типы триггеров. Обновим группу.

«Проектирование баз данных» - Нормализация. Этапы создания базы данных. Нормализованная база данных. Создание структуры базы данных и заполнение. Организация информации в табличную форму. Задание структуры базы данных. Работа с сохраненной базой данных. Таблица может быть: Хорошо нормализованной Плохо нормализованной. Плохо нормализованная таблица.

«Язык запросов SQL» - Агрегирующие функции. Применение односхемных операций РА. Примеры использования GROUP BY. Группировка по нескольким полям. Операции реляционной алгебры. Разность отношений. Пересечение отношений. Список сотрудников с указанием количества проектов. Использование фразы HAVING. Использование GROUP BY. Применение операции соединения.

«Большие объекты» - Упражнения. Поля большого размера. Вставка. Улучшенное двухуровневое разбиение. Запись. Удалить N байт. Древовидное представление. Удаление. Создание поля большого размера. Дескриптор поля большого размера. Алгоритм. Современные СУБД. Замечания. Большие объекты. Двухуровневое разбиение. Рисунок. Добавление в конец.

Работа с базами данных

11 презентаций о работе с базами данных


130 тем