Question 1

A set of pageviews requested by a single user from a Web server.&#10;A) index page&#10;B) common log&#10;C) session&#10;D) page frame

Accepted Answer

A session is a set of pageviews requested by a single user from a Web server. It tracks the user's activity on the website and can be used for various purposes such as personalization, tracking user behavior, and maintaining stateful information. The other options are not relevant to this definition.

Question 2

These can be used to help select a best subset of training data.&#10;A) domain resemblance scores&#10;B) class resemblance scores&#10;C) instance typicality scores&#10;D) standard deviation scores

Accepted Answer

C

Question 3

A data file that contains session information.&#10;A) cookie&#10;B) pageview&#10;C) page frame&#10;D) common log

Accepted Answer

D

Question 4

The automation of Web site adaptation involves creating and deleting&#10;A) index pages&#10;B) cookies&#10;C) pageviews&#10;D) clickstreams

Accepted Answer

The automation of website adaptation typically involves creating or deleting index pages, which are the main pages that users see when they first visit a website. The purpose of these pages is to provide users with an overview of the content and structure of the site, and to help them navigate to the specific pages they are interested in. While cookies, pageviews, and clickstreams can all be used in the process of website adaptation, they are typically not directly involved in the creation or deletion of index pages.

Question 5

A data mining algorithm designed to discover frequently accessed Web pages that occur in the same order.&#10;A) serial miner&#10;B) association rule miner&#10;C) sequence miner&#10;D) decision miner

Accepted Answer

The data mining algorithm described is used to discover frequently accessed web pages that occur in the same order. This is achieved through sequence mining. Option A, serial miner, is not a commonly used term in data mining. Option B, association rule miner, is typically used to discover co-occurrences of items in a transactional database. Option D, decision miner, is used to discover patterns in decision trees or decision rules.

Question 6

The training phase of a textual data mining process involves&#10;A) removing common words from a dictionary.&#10;B) creating an attribute dictionary.&#10;C) determining whether a document is about the topic under investigation.&#10;D) modifying an initially created attribute dictionary.

Accepted Answer

The training phase of a textual data mining process involves creating an attribute dictionary that contains the keywords or terms that are relevant to the topic under investigation. The attribute dictionary will be used to classify or categorize the documents based on their content. Removing common words from a dictionary and modifying an initially created attribute dictionary can be done in the pre-processing phase. Determining whether a document is about the topic under investigation is done in the classification phase.

Question 7

Which of the following is a fundamental difference between bagging and boosting?&#10;A) Bagging is used for supervised learning. Boosting is used with unsupervised clustering.&#10;B) Bagging gives varying weights to training instances. Boosting gives equal weight to all training instances.&#10;C) Bagging does not take the performance of previously built models into account when building a new model. With boosting each new model is built based upon the results of previous models.&#10;D) With boosting, each model has an equal weight in the classification of new instances. With bagging, individual models are given varying weights. Answers to Chapter 11 Questions&#10;Multiple Choice Questions

Accepted Answer

The fundamental difference between bagging and boosting is that bagging does not take the performance of previously built models into account when building a new model, whereas with boosting each new model is built based upon the results of previous models. Option A is incorrect because both bagging and boosting are used for supervised learning. Option B is incorrect because bagging can also give equal weights to all training instances. Option D is incorrect because in bagging, all models have equal weight in the classification of new instances.

Question 8

Which of the following problems is best solved using time-series analysis?&#10;A) Predict whether someone is a likely candidate for having a stroke.&#10;B) Determine if an individual should be given an unsecured loan.&#10;C) Develop a profile of a star athlete.&#10;D) Determine the likelihood that someone will terminate their cell phone contract.

Accepted Answer

Time-series analysis involves analyzing data points over time to identify trends and patterns. The problem of determining the likelihood that someone will terminate their cell phone contract requires analyzing historical data on customer behavior over time, making it the best choice for time-series analysis. The other problems listed do not necessarily involve analyzing data over time and may require different analytical approaches.

Question 9

At least eighty percent of the time spent on a Web-based data mining project is devoted to this.&#10;A) goal idenficiation&#10;B) data preparation&#10;C) data mining&#10;D) interpretation of results

Accepted Answer

Data preparation is the most time-consuming task in Web-based data mining projects. It involves collecting, cleaning, integrating, transforming, and selecting data from various sources for analysis. Goal identification, data mining, and interpretation of results are all important tasks, but they typically do not require as much time as data preparation.

Question 10

Usage profiles for Web-based personalization contain several&#10;A) pageviews&#10;B) clickstreams&#10;C) cookies&#10;D) session files

Accepted Answer

Pageviews are a fundamental component of usage profiles for web-based personalization, as they track the pages a user visits, which helps in understanding user preferences and behavior.

Deck 11: Specialized Techniques