Lectures 1,2,3

Information Storage and Retrieval
Chapter1:
Introduction to Information Retrieval Systems

OBJECTIVES
 Definition of Information Retrieval Systems
 Objectives of Information Retrieval Systems
 Functional Overview
 Relationship to Database Management Systems

Information Retrieval System Definition
 An Information Retrieval System is a
system that is capable of storage, retrieval,
and maintenance of information.
 Information in this context can be
composed of text (including numeric and
date data), images, audio, video and other
multi-media objects.
 Techniques are beginning to emerge to
search these other media types.

Gauge of an IR System
 An Information Retrieval System consists of a
software program that facilitates a user in finding
the information file user needs.
 The gauge of success of an information system
is how well it can minimize the overhead for a
user to find the needed information.
 Overhead from a user's perspective is tile time
required to find tile information needed,
excluding the time for actually reading the
relevant data. Thus search composition, search
execution, and reading non-relevant items are all
aspects of information retrieval overhead.

What is an Item?
 The term "item" is used to represent the smallest
complete textual unit that is processed and
manipulated by the system.
 The definition of item varies by how a specific
source treats information. A complete document,
such as a book, newspaper or magazine could
be an item. At other times each chapter, or article
may be defined as an item.
 As sources vary and systems include more
complex processing, an item may address even
lower levels of abstraction such as a contiguous
passage of text or a paragraph.

Objectives of an IR System
 The general objective of an Information
Retrieval System is to minimize the overhead
of a user locating needed information.
 Overhead can be expressed as the time a
user spends in all of the steps leading to
reading an item containing the needed
information (e.g., query generation, query
execution, scanning results of query to select
items to read, reading non-relevant items).

Measures associates with IR systems
 The two major measures
commonly associated with
information systems are
precision and recall.

• When a user decides to issue a search
looking for information on a topic, the total
database is logically divided into four
segments

Cont.
 Relevant items are those documents that
contain information that helps the searcher
in answering his question.
 Non-relevant items are those items that
do not provide any directly useful
information.
 There are two possibilities with respect to
each item: it can be retrieved or not
retrieved by the user's query.

Cont.
Where:
 Number_Possible_Relevant are the
number of relevant items in the database.
 Number_Total__Retieved is the total
number of items retrieved from the query.
 Number_Retrieved_Relevant is the
number of items retrieved that are relevant
to the user's search need.

Cont.
 Precision measures one aspect of
information retrieval overhead for a user
associated with a particular search.
 If a search has a 85 per cent precision,
then 15 per cent of the user effort is
overhead reviewing non-relevant items.
 Recall gauges how well a system
processing a particular query is able to
retrieve the relevant items that the user is
interested in seeing.

Ideal Precision and Recall
 Figure 1.2a shows the values of precision and recall as the
number of items retrieved increases, under an optimum query
where every returned item is relevant. There are "N" relevant
items in the database.
 In Figure 1.2a the basic properties of precision (solid line) and
recall (dashed line) can be observed.
 Precision starts off at 100 per cent and maintains that value as
long as relevant items are retrieved.
 Recall starts off close to zero and increases as long as
relevant items are retrieved until all possible relevant items
have been retrieved.
 Once all "N" relevant items have been retrieved, the only items
being retrieved are non-relevant. Precision is directly affected
by retrieval of non-relevant items and drops to a number close
to zero. Recall is not effected by retrieval of non-relevant items

Cont.
 The first objective of an Information Retrieval
System is support of user search generation.
 Natural languages suffer from word
ambiguities such as homographs and use of
acronyms that allow the same word to have
multiple meanings (e.g., the word "field“).
 Disambiguation techniques exist but introduce
significant system overhead in processing
power and extended search times and often
require interaction with the user.

Cont.
 Many users have trouble in generating a good
search statement. The typical user does not
have significant experience with nor even the
aptitude for Boolean logic statements.
 Quite often the user is not an expert in the
area that is being searched and lacks domain
specific vocabulary unique to that particular
subject area (Search begins with a general
concept, a limited knowledge of the
vocabulary associated with a particular area).

Cont.
 Even when the user is an expert in the
area being searched, the ability to select
the proper search terms is constrained by
lack of knowledge of the author's
vocabulary.
 Thus, an Information Retrieval System
must provide tools to help overcome the
search specification problems discussed
above.

Cont.
 An objective of an information system is to
present the search results in a format that
facilitates the user in determining relevant
items.
 Historically data has been presented in an
order dictated by how it was physically
stored. Typically, this is in arrival to the
system order, thereby always displaying
the results of a search sorted by time. For
those users interested in current events
this is useful.

Cont.
 The new Information Retrieval Systems
provide functions that provide the results of
a query in order of potential relevance to
the user.
 Even more sophisticated techniques use
item clustering and link analysis to provide
additional item selection insights.

IR Systems Functional Overview
 A total Information Storage and Retrieval
System is composed of four major functional
processes:
1. Item Normalization,
2. Selective Dissemination of Information (i.e.,
"Mail"),
3. Archival Document Database Search, and
4. An Index Database Search.
 Commercial systems have not integrated
these capabilities into a single system but
supply them as independent capabilities.

1. Item Normalization
 Normalize the incoming items to a
standard format.
 Standardizing the input takes the different
external formats of input data and performs
the translation to the formats acceptable to
the system.
 A system may have a single format for all
items or allow multiple formats.

Cont.
 The next process is to parse the item into logical
sub-divisions that have meaning to the user. This
process, called "Zoning," is visible to the user
and used to increase the precision of a search
and optimize the display.
 An item is subdivided into zones, which may be
hierarchical (Title, Author, Abstract, Main Text,
Conclusion, and References).
 The zoning information is passed to the
processing token identification operation to store
the information, allowing searches to be
restricted to a specific zone.

Cont.
 Once the standardization and zoning has
been completed, information (i.e., words)
that are used in the search process need
to be identified in the item.
 The first step in identification of a
processing token consists of determining a
word. Systems determine words by
dividing input symbols into three classes:
valid word symbols, inter-word symbols,
and special processing symbols.

Cont.
 A word is defined as a contiguous set of
word symbols bounded by inter-word
symbols.
 Examples of word symbols are alphabetic
characters and numbers.
 Examples of possible inter-word symbols
are blanks, periods and semicolons.

Cont.
 Next, a Stop List/Algorithm is applied to the list of
potential processing tokens.
 The objective of the Stop function is to save
system resources by eliminating from the set of
searchable processing tokens those that have
little value to the system.
 Stop Lists are commonly found in most systems
and consist of words (processing tokens) whose
frequency and/or semantic use make them of no
value as a searchable token.
 (e.g., "the"), have no search value and are not a
useful part of a user's query.

Cont.
 The next step in finalizing on processing
tokens is identification of any specific word
characteristics.
 The characteristic is used in systems to
assist in disambiguation of a particular
word.
 Morphological analysis of the processing
token's part of speech is included here.

Cont.
 Once the potential processing token has been
identified and characterized, most systems apply
stemming algorithms to normalize the token to a
standard semantic representation.
 The decision to perform stemming is a trade off
between precision of a search (i.e., finding
exactly what the query specifies) versus
standardization to reduce system overhead in
expanding a search term to similar token
representations with a potential increase in
recall.
 The amount of stemming that is applied can lead
to retrieval of many non-relevant items.

2. Selective Dissemination of Information
 (Mail) Process provides tile capability to
dynamically compare newly received items in the
information system against standing statements
of interest of users and deliver the item to those
users whose statement of interest matches the
contents of the item.
 The Mail process is composed of the search
process, user statements of interest (Profiles)
and user mail files.
 When the search statement is satisfied, the item
is placed in the Mail File(s) associated with the
profile.

2. Selective Dissemination of Information
Cont.
 As each item is received, it is processed against
every user's profile. A profile contains a typically
broad search statement along with a list of user
mail files that will receive the document if the
search statement in the profile is satisfied.
 User search profiles are different than ad hoc
queries in that they contain significantly more
search terms (10 to 100 times more terms) and
cover a wider range of interests.
 These profiles define all the areas in which a
user is interested versus an ad hoc query which
is frequently focused to answer a specific
question.

3. Document Database Search
 The Document Database Search process is
composed of the search process, user entered
queries (typically ad hoc queries) and the
document database which contains all items that
have been received, processed and stored by
the system.
 Any search for information that has already been
processed into the system can be considered a
"retrospective" search for information.
 Queries differ from profiles in that they are
typically short and focused on a specific area of
interest.

4. Index Database Search
 When an item is determined to be of
interest, a user may want to save it for
future reference. This is in effect filing it.
 In an information system this is
accomplished via the index process. In this
process the user can logically store an
item in a file along with additional index
terms and descriptive text the user wants
to associate with the item.

 The Index Database Search Process
provides the capability to create indexes
and search them.
 The user may search the index and
retrieve the index and/or the document it
references.
 The system also provides the capability to
search the index and then search the items
referenced by the index records that
satisfied the index portion of the query.
This is called a combined file search.

 There are two classes of index files: Public
and Private Index files.
 Every user can have one or more Private
Index files leading to a very large number of
files. Each Private Index file references only a
small subset of the total number of items in
the Document Database.
 Public Index files are maintained by
professional library services personnel and
typically index every item in the Document
Database.

Relationship to Database Management
Systems
 1. An Information Retrieval System is software
that has the features and functions required to
manipulate "information" items versus a DBMS
that is optimized to handle "structured" data.
Information is fuzzy text.
 2. Structured data is well defined data (facts)
typically represented by tables. There is a
semantic description associated with each
attribute within a table that well defines that
attribute. On the other hand, if two different
people generate an abstract for the same item,
they can be different.

Relationship to Database Management
Systems
 3. With structured data a user enters a specific
request and the results returned provide the user
with the desired information. The results are
frequently tabulated and presented in a report
format for ease of use. In contrast, a search of
"information" items has a high probability of not
finding all the items a user is looking for. The
user has to refine his search to locate additional
items of interest. This process is called "iterative
search.“
 From a practical standpoint, the integration of
DBMS's and Information Retrieval Systems is
very important.

Information Retrieval System
Capabilities
Chapter 2

Objectives
 Discussing the major functions that
are available in an Information
Retrieval System.
 Search and browse capabilities are
crucial to assist the user in locating
relevant items.

Search Capabilities
 The objective of the search capability is to allow for a
mapping between a user's specified need and the
items in the information database that will answer
that need.
 “Weighting" of search terms holds significant
potential for assisting in the location and ranking of
relevant items.
 E.g. Find articles that discuss data mining(.9) or data
warehouses(.3).
 the system would recognize in its importance ranking
and item selection process that data mining are far
more important than items discussing data
warehouses.

1. Boolean Logic
 Boolean logic allows a user to logically relate
multiple concepts together to define what information
is needed. The typical Boolean operators are AND,
OR, and NOT.
 Placing portions of the search statement in
parentheses are used to overtly specify the order of
Boolean operations (i.e., nesting function). If
parentheses are not used, the system follows a
default precedence ordering of operations (e.g.,

2. Proximity
 Proximity is used to restrict the distance allowed within an item
between two search terms.
 The semantic concept is that the closer two terms are found in
a text the more likely they are related in the description of a
particular concept.
 Proximity is used to increase the precision of a search.
 If the terms COMPUTER and DESIGN are found within a few
words of each other then the item is more likely to be
discussing the design of computers than if the words are
paragraphs apart.

2. Proximity
 TERM1 within "m . . . . units" of TERM2
 The distance operator "m" is an integer number and units
are in Characters, Words, Sentences, or Paragraphs.
 A special case of the Proximity operator is the Adjacent
(ADJ) operator that normally has a distance operator of
one and a forward only direction.
 Another special case is where the distance is set to zero
meaning within the same semantic unit.

3. Contiguous Word Phrases
 A Contiguous Word Phrase (CWP) is both a way of
specifying a query term and a special search operator. A
Contiguous Word Phrase is two or more words that are
treated as a single semantic unit.
 An example of a CWP is "United States of America." It is
four words that specify a search term representing a
single specific semantic concept (a country) that can be
used with any of the operators discussed above.
 Thus a query could specify "manufacturing" AND "United
States of America" which returns any item that contains
the word "manufacturing" and the contiguous words
"United States of America”.
 A contiguous word phrase also acts like a special search
operator that is similar to the proximity (Adjacency)
operator but allows for additional specificity.

4. Fuzzy Searches
 Fuzzy Searches provide the capability to locate spellings of
words that are similar to the entered search term. This function
is primarily used to compensate for errors in spelling of words.
 Fuzzy searching increases recall at the expense of decreasing
precision.
 A Fuzzy Search on the term "computer" would automatically
include the following words from the information database:
"computer”, "compiter," "conputer," "computter," "compute."
 An additional enhancement may lookup the proposed
alternative spelling and if it is a valid word with a different
meaning, include it in the search with a low ranking or not
include it at all (e.g., "commuter").
 In the process of expanding a query term fuzzy searching
includes other terms that have similar spellings, giving more
weight (in systems that rank output) to words in the database
that have similar word lengths and position of the characters
as the entered term.

5. Term Masking
 Term masking is the ability to expand a query term by
masking a portion of the term and accepting as valid any
processing token that maps to the unmasked portion of
the term. The value of term masking is much higher in
systems that do not perform stemming or only provide a
very simple stemming algorithm.
 There are two types of search term masking: fixed length
and variable length.
 Fixed length masking is a single position mask. It masks
out any symbol in a particular position or the lack of that
position in a word.
 Variable length "don't cares" allows masking of any
number of characters within a processing token.

5. Term Masking
(Variable Length)

6. Numeric and Date Ranges
 Term masking is useful when applied to words, but does
not work for finding ranges of numbers or numeric dates.
 To find numbers larger than "125”, using a term "125*"
will not find any number except those that begin with the
digits "125.“
 A user could enter inclusive (e.g., "125-425" or "4/2/93-
5/2/95" for numbers and dates) to infinite ranges (">125“,
"<=233“, representing "Greater Than" or "Less Than” or
“Equal") as part of a query.

7. Concept/Thesaurus Expansion
 Associated with both Boolean and Natural Language
Queries is the ability to expand the search terms via
Thesaurus or Concept Class database reference tool.
 A Thesaurus is typically a one-level or two-level
expansion of a term to other terms that are similar in
meaning.
 A Concept Class is a tree structure that expands each
meaning of a word into potential concepts that are related
to the initial term.

7. Concept/Thesaurus Expansion

8. Natural Language Queries
 Natural Language Queries allow a user to enter a
prose statement that describes the information that
the user wants to find.
 The longer the prose, the more accurate file results
returned. The most difficult logic case associated
with Natural Language Queries is the ability to
specify negation in the search statement and have
the system recognize it as negation.

 An example of a Natural Language Query is:
 Find for me all the items that discuss databases and
current attempts in database applications. Include all
items that discuss Microsoft trials in the development
process. Do not include items about relational
databases.

 This usage pattern is important because sentence
fragments make morphological analysis of the natural
language query difficult and may limit the system's ability to
perform term disambiguation (e.g., understand which
meaning of a word is meant).
 Natural language interfaces improve the recall of systems
with a decrease in precision when negation is required.

Browse Capabilities
 Browse capabilities provide the user with the
capability to determine which items are of interest
and select those to be displayed.
 There are two ways of displaying a summary of the
items that are associated with a query: line item
status and data visualization.
 If searches resulted in high precision, then the
importance of the browse capabilities would be
lessened.
 Since searches return many items that are not
relevant to the user's information need, browse
capabilities can assist the user in focusing on items
that have the highest likelihood in meeting his need.

1. Ranking
 Hits are retrieved in either a sorted order (e.g., sort by Title) or in
time order from the newest to the oldest item.
 With the introduction of ranking based upon predicted relevance
values, the status summary displays the relevance score
associated with the item along with a brief descriptor of the item
(usually both fit on one display screen line).
 The relevance score is an estimate of the search system on how
closely the item satisfies the search statement. Typically
relevance scores are normalized to a value between 0.0 and 1.0.
The highest value of 1.0 is interpreted that the system is sure that
the item is relevant to the search statement.

1. Ranking
 Practically, systems have a default minimum value which the
user can modify that stops returning items that have a
relevance value below the specified value.
 Presenting the actual relevance number seems to be more
confusing to the user than presenting a category that the
number falls in.
 For example, some systems create relevance categories and
indicate, by displaying items in different colors, which category
an item belongs to. Other systems uses a nomenclature such
as High, Medium High, Medium, Low, and Non-relevant. The
color technique removes the need for written indication of an

1. Ranking
 Rather than limiting the number of items that can be assessed
by the number of lines on a screen, other graphical
visualization techniques showing the relevance relationships
of the hit items can be used.
 For example, a two or three dimensional graph can be
displayed where points on the graph represent items and the
location of the points represent their relative relationship
between each other and the user's query.
 This technique allows a user to see the clustering of items by
topics and browse through a cluster or move to another topical
cluster.

2. Zoning
 The user wants to see the minimum information needed
to determine if the item is relevant.
 Limited display screen sizes require selectability of what
portions of an item a user needs to see to make the
relevance determination.
 For example, display of the Title and Abstract may be
sufficient information for a user to predict the potential
relevance of an item. Limiting the display of each item to
these two zones allows multiple items to be displayed on
a single display screen.
 This makes maximum use of tile speed of the user's
cognitive process in scanning the single image and
understanding the potential relevance of the multiple
items on the screen.

3. Highlighting
 Lets the user quickly focus on the potentially relevant
parts of the text to scan for item relevance.
 Most systems allow the display of an item to begin with
the first highlight within tile item and allow subsequent
jumping to the next highlight.
 Another capability, which is gaining strong acceptance, is
for the system to determine the passage in the document
most relevant to the query and position the browse to
start at that passage.
 Using Natural Language Processing, and automatic
expansion of terms via thesauri; highlighting loses some
of its value.
 The terms being highlighted that caused a particular item
to be returned may not have direct or obvious mapping to
any of the search terms entered.

Miscellaneous Capabilities
 There are many additional functions that facilitate the
user's ability to input queries, reducing the time it
takes to generate the queries, and reducing a priori
the probability of entering a poor query.

1. Vocabulary Browse
 The capability to display in alphabetical sorted order
words from the document database.
 The user can enter a word or word fragment and the
system will begin to display file dictionary around the
entered text.
 It helps the user determine the impact of using a
fixed or variable length mask on a search term and
potential mis-spellings.
 The user can determine that entering the search
term "compul*" in effect is searching for "compulsion"
or compulsive" or "compulsory."

2. Iterative Search and Search History Log
 The process of refining the results of a previous
search to focus on relevant items is called iterative
search.
 To facilitate locating previous searches as starting
points for new searches, search history logs are
available.
 The search history log is the capability to display all
the previous searches that were executed during the
current session.

3. Canned Query
 The capability to name a query and store it to be
retrieved and executed during a later user session is
called canned or stored queries.
 A canned query focuses on the user's general area
of interest one time and then retrieve it to add
additional search criteria to retrieve data that is
currently needed.
 Queries that start with a canned query are
significantly larger than ad hoc queries.

Lectures 1,2,3

Recommended

More Related Content

What's hot (20)

Similar to Lectures 1,2,3 (20)

Recently uploaded (16)

Lectures 1,2,3