Please read this disclaimer before proceeding:

This document is confidential and intended solely for the educational purpose of
RMK Group of Educational Institutions. If you have received this document
through email in error, please notify the system manager. This document
contains proprietary information and is intended only to the respective group /
learning community as intended. If you are not the addressee you should not
disseminate, distribute or copy through e-mail. Please notify the sender
immediately by e-mail if you have received this document by mistake and delete
this document from your system. If you are not the intended recipient you are
notified that disclosing, copying, distributing or taking any action in reliance on
the contents of this information is strictly prohibited.
 CS8080

INFORMATION
RETRIEVAL
TECHNIQUES
Department: CSE UNIT I

Batch/Year: 2020-24 / IV YEAR

Created by:
Dr S.SRINIVASAN, Professor,
CSE Department, RMDEC

Date: 06.11.2023
 Table of Contents

Course Objectives
Pre Requisites

Syllabus

Course outcomes
CO- PO/PSO Mapping

Lecture Plan

Activity based learning

Unit I:

Lecture Notes
Lecture Notes – Links to Videos
Lecture Notes – e book reference
Lecture Notes – PPTs

Lecture Notes - Quiz

Lecture Notes - References

Assignments
Part A Q & A (with K level and CO)

Part B Qs (with K level and CO)

Supportive online Certification courses

Real time Applications in day to day life and to Industry

Contents beyond the Syllabus

Assessment Schedule
Prescribed Text Books & Reference Books

Mini Project suggestions

 Course Objectives

Understand the basics of Information Retrieval.

Use an open source search engine framework and explore its

capabilities.

Apply appropriate method of classification or clustering.

Design and implement innovative features in a search engine.

Design and implement a recommender system.

 Pre Requisites

CS8492 Database Management Systems

CS8691 Artificial Intelligence

CS8391 Data Structures

 Syllabus
CS8080 INFORMATION RETRIEVAL TECHNIQUES LTPC
3003

UNIT I INTRODUCTION 9
Information Retrieval – Early Developments – The IR Problem – The User‗s Task –
Information versus Data Retrieval - The IR System – The Software Architecture of
the IR System – The Retrieval and Ranking Processes - The Web – The e-Publishing
Era – How the web changed Search – Practical Issues on the Web – How People
Search – Search Interfaces Today – Visualization in Search Interfaces.

UNIT II MODELING AND RETRIEVAL EVALUATION 9

Basic IR Models - Boolean Model - TF-IDF (Term Frequency/Inverse Document
Frequency) Weighting - Vector Model – Probabilistic Model – Latent Semantic
Indexing Model – Neural Network Model – Retrieval Evaluation – Retrieval Metrics –
Precision and Recall – Reference Collection – User-based Evaluation – Relevance
Feedback and Query Expansion – Explicit Relevance Feedback.

UNIT III TEXT CLASSIFICATION AND CLUSTERING 9

A Characterization of Text Classification – Unsupervised Algorithms: Clustering –
Naïve Text Classification – Supervised Algorithms – Decision Tree – k-NN Classifier –
SVM Classifier – Feature Selection or Dimensionality Reduction – Evaluation metrics
– Accuracy and Error – Organizing the classes – Indexing and Searching – Inverted
Indexes – Sequential Searching – Multi-dimensional Indexing.

UNIT IV WEB RETRIEVAL AND WEB CRAWLING 9

The Web – Search Engine Architectures – Cluster based Architecture – Distributed
Architectures – Search Engine Ranking – Link based Ranking – Simple Ranking
Functions – Learning to Rank – Evaluations -- Search Engine Ranking – Search
Engine User Interaction – Browsing – Applications of a Web Crawler – Taxonomy –
Architecture and Implementation – Scheduling Algorithms – Evaluation.

UNIT V RECOMMENDER SYSTEM 9

Recommender Systems Functions – Data and Knowledge Sources –
Recommendation Techniques – Basics of Content-based Recommender Systems –
High Level Architecture – Advantages and Drawbacks of Content-based Filtering –
Collaborative Filtering – Matrix factorization models – Neighborhood models.

TOTAL: 45 PERIODS
 Course outcomes
Tounderstand the basics of Information Retrieval

Tounderstand machine learning techniques for text classification and clustering.

Tounderstand various search engine system operations.

Tolearn different techniques of recommender system

 CO- PO/PSO Mapping

CO PO PO PO PO PO PO PO PO PO PO PO PO PS PS PS

1 2 3 4 5 6 7 8 9 10 11 12 O1 O2 O3

1 3 3 3 - - - - - - - - - 3 - -

2 3 2 2 - - - - - - - - - 3 - -

3 3 3 3 - - - - - - - - - 3 - -

4 3 3 3 - - - - - - - - - 3 - -
 Unit V

RECOMMENDER
SYSTEM
 Lecture Plan
per
Ses No. Proposed Actual tai Taxon Mode
Topics to be date nin
sion of Lecture omy of
No. covered Per Date g level Deliver
iod CO y
s
Recommender Systems
Functions
1 1 1 K1 PPT

Data and Knowledge

2 Sources 1
1 K1 PPT
Recommendation
Techniques
3 1 1 K1 PPT
Basics of Content-based
4 Recommender Systems 1 1 K1 PPT

High Level Architecture

5 1 1 K1 PPT

Advantages and
Drawbacks of Content-
6 based Filtering 1 1 K1 PPT

Collaborative Filtering
7 1 1 K2 PPT

Matrix factorization
8 models 1 1 K2 PPT

Neighborhood models.
9 1 1 K2 PPT
Recommender Systems
Function
Introduction
Most of today’s internet businesses deeply root their success in the ability to provide
users with strongly personalized experiences

Definition
Recommender Systems (RSs) are software tools and techniques providing
suggestions for items to be of use to a user

Example: Amazon Recommendations

The value of recommendations
Netflix: 2/3 of the movies watched are recommended
Google News: recommendations generate 38% more clickthrough
Amazon: 35% sales from recommendations

Choicestream: 28% of the people would buy more music if they found what they
liked.
Recommender Systems Function
Its primary function is to locate documents that are relevant to the user’s
information need, but it can also be used to check the importance of a Web page or
to discover the various usages of a word in a collection of documents.

Find Some Good Items: Recommend to a user some items as a ranked list along
with predictions of how much the user would like them (e.g., on a one- to fivestar
scale). This is the main recommendation task that many commercial systems
address Some systems do not show the predicted rating.

• Find all good items: Recommend all the items that can satisfy some user needs. In
such cases it is insufficient to just find some good items. This is especially true when
the number of items is relatively small or when the RS is mission-critical, such as in
medical or financial applications. In these situations, in addition to the benefit
derived from carefully examining all the possibilities, the user may also benefit from
the RS ranking of these items or from additional explanations that the RS generates.
Recommender Systems
Function
•Annotation in context: Given an existing context, e.g., a list of items, emphasize
some of them depending on the user’s long-term preferences. For example, a TV
recommender system might annotate which TV shows displayed in the electronic
program guide (EPG) are worth watching
. • Recommend a sequence: Instead of focusing on the generation of a single
recommendation, the idea is to recommend a sequence of items that is pleasing as
a whole. Typical examples include recommending a TV series; a book on RSs after
having recommended a book on data mining; or a compilation of musical tracks
Recommend a bundle: Suggest a group of items that fits well together. For instance
a travel plan may be composed of various attractions, destinations, and
accommodation services that are located in a delimited area. From the point of view
of the user these various alternatives can be considered and selected as a single
travel destination.
Just browsing: In this task, the user browses the catalog without any imminent
intention of purchasing an item. The task of the recommender is to help the user to
browse the items that are more likely to fall within the scope of the user’s interests
for that specific browsing session. This is a task that has been also supported by
adaptive hypermedia techniques.

• Find credible recommender: Some users do not trust recommender systems

thus they play with them to see how good they are in making recommendations.
Hence, some system may also offer specific functions to let the users test its
behavior in addition to those just required for obtaining recommendations.

• Improve the profile: This relates to the capability of the user to provide (input)
information to the recommender system about what he likes and dislikes. This is a
fundamental task that is strictly necessary to provide personalized
recommendations. If the system has no specific knowledge about the active user
then it can only provide him with the same recommendations that would be
delivered to an “average” user.
Recommender Systems
Function
•Express self: Some users may not care about the recommendations at all. Rather,
what it is important to them is that they be allowed to contribute with their ratings
and express their opinions and beliefs. The user satisfaction for that activity can still
act as a leverage for holding the user tightly to the application (as we mentioned
above in discussing the service provider’s motivations).
•Help others: Some users are happy to contribute with information, e.g., their
evaluation of items (ratings), because they believe that the community benefits from
their contribution. This could be a major motivation for entering information into a
recommender system that is not used routinely. For instance, with a car RS, a user,
who has already bought her new car is aware that the rating entered in the system
is more likely to be useful for other users rather than for the next time she will buy a
car
. • Influence others: In Web-based RSs, there are users whose main goal is to
explicitly influence other users into purchasing particular products. As a matter of
fact, there are also some malicious users that may use the system just to promote
or penalize certain items

Applications
Increase the number of items sold
Sell more diverse items
Increase the user satisfaction
Increase user fidelity
Better understand what the user wants
Data and Knowledge Sources
• RSs are information processing systems that actively gather various kinds of
data in order to build their recommendations.
• Data is primarily about the items to suggest and the users who will receive these
recommendations. But, since the data and knowledge sources available for
recommender systems can be very diverse, ultimately, whether they can be
exploited or not depends on the recommendation technique.
• In general, there are recommendation techniques that are knowledge poor, i.e.,
they use very simple and basic data, such as user ratings/evaluations .

• Other techniques are much more knowledge dependent, e.g., using ontological

descriptions of the users or the items or constraints or social relations and

activities of the users In any case, as a general classification, data used by RSs
refers to three kinds of objects: items, users, and transactions, i.e., relations
between users and items. Items.

• Items are the objects that are recommended. Items may be characterized by

their complexity and their value or utility. The value of an item may be positive if
the item is useful for the user, or negative if the item is not appropriate and the
user made a wrong decision when selecting it.
• We note that when a user is acquiring an item she will always incur in a cost,
which includes the cognitive cost of searching for the item and the real monetary
cost eventually paid for the item. For instance, the designer of a news RS must
take into account the complexity of a news item, i.e., its structure, the textual
representation, and the time-dependent importance of any news item. But, at the
same time, the RS designer must understand that even if the user is not paying
for reading news, there is always a cognitive cost associated to searching and
reading news items.
 Data and Knowledge Sources

• If a selected item is relevant for the user this cost is dominated by the benefit of
having acquired a useful information, whereas if the item is not relevant the net
value of that item for the user, and its recommendation, is negative. In other
domains, e.g., cars, or financial investments, the true monetary cost of the items
becomes an important element to consider when selecting the most appropriate
recommendation approach. Items with low complexity and value are: news, Web
pages, books, CDs, movies. Items with larger complexity and value are: digital
cameras, mobile phones, PCs, etc.
• The most complex items that have been considered are insurance policies, financial
investments, travels, jobs.

• RSs, according to their core technology, can use a range of properties and features
of the items. For example in a movie recommender system, the genre (such as
comedy, thriller, etc.), as well as the director, and actors can be used to describe a
movie and to learn how the utility of an item depends on its features. Items can be
represented using various information and representation approaches, e.g., in a
minimalist way as a single id code, or in a richer form, as a set of attributes, but
even as a concept in an ontological representation of the domain.

• Users:
• Users of a RS, as mentioned above, may have very diverse goals and
characteristics. In order to personalize the recommendations and the human-
computer interaction, RSs exploit a range of information about the users. This
information can be structured in various ways and again the selection of what
information to model depends on the recommendation technique. For instance, in
collaborative filtering, users are modeled as a simple list containing the ratings
provided by the user for some items. In a demographic RS, sociodemographic
attributes such as age, gender, profession, and education, are used. User data is
said to constitute the user model.
 Data and Knowledge Sources

The user model profiles the user, i.e., encodes her preferences and needs. Various
user modeling approaches have been used and, in a certain sense, a RS can be
viewed as a tool that generates recommendations by building and exploiting user
models. Since no personalization is possible without a convenient user model, unless
the recommendation is non-personalized, as in the top-10 selection, the user model
will always play a central role. For instance, considering again a collaborative filtering
approach, the user is either profiled directly by its ratings to items or, using these
ratings, the system derives a vector of factor values, where users differ in how each
factor weights in their model.
Users can also be described by their behavior pattern data, for example, site
browsing patterns (in a Web-based recommender system), or travel search patterns
(in a travel recommender system). Moreover, user data may include relations
between users such as the trust level of these relations between users. A RS might
utilize this information to recommend items to users that were preferred by similar or
trusted users.

Transactions.
We generically refer to a transaction as a recorded interaction between a user and
the RS. Transactions are log-like data that store important information generated
during the human-computer interaction and which are useful for the recommendation

generation algorithm that the system is using.

For instance, a transaction log may contain a reference to the item selected by the
user and a description of the context (e.g., the user goal/query) for that particular
recommendation. If available, that transaction may also include an explicit feedback
the user has provided, such as the rating for the selected item. In fact, ratings are
the most popular form of transaction data that a RS collects. These ratings may be
collected explicitly or implicitly. In the explicit collection of ratings, the user is asked
to provide her opinion about an item on a rating scale.
 Data and Knowledge Sources

Ratings can take on a variety of forms:

• Numerical ratings such as the 1-5 stars provided in the book recommender
associated with Amazon.com.
• Ordinal ratings, such as “strongly agree, agree, neutral, disagree, strongly
disagree” where the user is asked to select the term that best indicates her opinion
regarding an item (usually via questionnaire).
• Binary ratings that model choices in which the user is simply asked to decide if a
certain item is good or bad.
• Unary ratings can indicate that a user has observed or purchased an item, or
otherwise rated the item positively. In such cases, the absence of a rating indicates
that we have no information relating the user to the item (perhaps she purchased
the item somewhere else)
In transactions collecting implicit ratings, the system aims to infer the users opinion
based on the user’s actions. For example, if a user enters the keyword “Yoga” at
Amazon.com she will be provided with a long list of books. In return, the user may
click on a certain book on the list in order to receive additional information.
At this point, the system may infer that the user is somewhat interested in that book.
In conversational systems, i.e., systems that support an interactive process, the
transaction model is more refined. In these systems user requests alternate with
system actions.

That is, the user may request a recommendation and the system may produce a
suggestion list. But it can also request additional user preferences to provide the user
with better results. Here, in the transaction model, the system collects the various
requests-responses, and may eventually learn to modify its interaction strategy by
observing the outcome of the recommendation process.
 Data and Knowledge Sources

Ratings can take on a variety of forms:

• Numerical ratings such as the 1-5 stars provided in the book recommender
associated with Amazon.com.
• Ordinal ratings, such as “strongly agree, agree, neutral, disagree, strongly
disagree” where the user is asked to select the term that best indicates her opinion
regarding an item (usually via questionnaire).
• Binary ratings that model choices in which the user is simply asked to decide if a
certain item is good or bad.
• Unary ratings can indicate that a user has observed or purchased an item, or
otherwise rated the item positively. In such cases, the absence of a rating indicates
that we have no information relating the user to the item (perhaps she purchased
the item somewhere else)
In transactions collecting implicit ratings, the system aims to infer the users opinion
based on the user’s actions. For example, if a user enters the keyword “Yoga” at
Amazon.com she will be provided with a long list of books. In return, the user may
click on a certain book on the list in order to receive additional information.
At this point, the system may infer that the user is somewhat interested in that book.
In conversational systems, i.e., systems that support an interactive process, the
transaction model is more refined. In these systems user requests alternate with
system actions.

That is, the user may request a recommendation and the system may produce a
suggestion list. But it can also request additional user preferences to provide the user
with better results. Here, in the transaction model, the system collects the various
requests-responses, and may eventually learn to modify its interaction strategy by
observing the outcome of the recommendation process.
 Recommendation Techniques

• In order to implement its core function, identifying the useful items for the user, a
RS must predict that an item is worth recommending. In order to do this, the
system must be able to predict the utility of some of them, or at least compare
the utility of some items, and then decide what items to recommend based on this
comparison. The prediction step may not be explicit in the recommendation
algorithm but we can still apply this unifying model to describe the general role of
a RS.

• To illustrate the prediction step of a RS, consider, for instance, a simple,

nonpersonalized, recommendation algorithm that recommends just the most
popular songs.

• The rationale for using this approach is that in absence of more precise
information about the user’s preferences, a popular song, i.e., something that is
liked (high utility) by many users, will also be probably liked by a generic user, at
least more than another randomly selected song. Hence the utility of these
popular songs is predicted to be reasonably high for this generic user
•
• This view of the core recommendation computation as the prediction of the utility
of an item for a user. They model this degree of utility of the user u for the item i
as a (real valued) function R(u,i), as is normally done in collaborative filtering by
considering the ratings of users for items. Then the fundamental task of a
collaborative filtering RS is to predict the value of R over pairs of users and items,
i.e., to compute 𝑅෡̈ (u,i), where we denote with 𝑅෡̈ the estimation, computed by
the RS, of the true function R. Consequently, having computed this prediction for
the active user u on a set of items, i.e., 𝑅෡ ̈ (u,i1),...,𝑅෡ ̈ (u,iN) the system will
recommend the items ij1 ,...,ijK (K ≤ N) with the largest predicted utility. K is
typically a small number, i.e., much smaller than the cardinality of the item data
set or the items on which a user utility prediction can be computed, i.e., RSs
“filter” the items that are recommended to users.
 Recommendation Techniques

Some recommender systems do not fully estimate the utility before making a
recommendation but they may apply some heuristics to hypothesize that an item is of
use to a user. This is typical, for instance, in knowledge-based systems. These utility
predictions are computed with specific algorithms and use various kind of knowledge
about users, items, and the utility function itself.

Six different classes of recommendation approaches:

1.Content-based: The system learns to recommend items that are similar to the ones
that the user liked in the past. The similarity of items is calculated based on the
features associated with the compared items. For example, if a user has positively
rated a movie that belongs to the comedy genre, then the system can learn to
recommend other movies from this genre

2. Collaborative filtering: The simplest and original implementation of this approach

recommends to the active user the items that other users with similar tastes liked in
the past. The similarity in taste of two users is calculated based on the similarity in
the rating history of the users. This is the reason why refers to collaborative filtering
as “people-to-people correlation.”
3.Demographic: This type of system recommends items based on the demographic
profile of the user. The assumption is that different recommendations should be
generated for different demographic niches. Many Web sites adopt simple and
effective personalization solutions based on demographics. For example, users are
dispatched to particular Web sites based on their language or country.
4.Knowledge-based: Knowledge-based systems recommend items based on specific
domain knowledge about how certain item features meet users needs and
preferences and, ultimately, how the item is useful for the user. Notable knowledge
based recommender systems are case-based. In these systems a similarity function
estimates how much the user needs (problem description) match the
recommendations (solutions of the problem). Here the similarity score can be directly

interpreted as the utility of the recommendation for the user.

 Recommendation Techniques

5.Community-based: This type of system recommends items based on the

preferences of the users friends. This technique follows the epigram “Tell me who
your friends are, and I will tell you who you are”.. Evidence suggests that people
tend to rely more on recommendations from their friends than on recommendations
from similar but anonymous individuals.
6.Hybrid recommender systems: These RSs are based on the combination of the
above mentioned techniques. A hybrid system combining techniques A and B tries to
use the advantages of A to fix the disadvantages of B. For instance, CF methods
suffer from new-item problems, i.e., they cannot recommend items that have no
ratings. This does not limit content-based approaches since the prediction for new
items is based on their description (features) that are typically easily available. Given
two (or more) basic RSs techniques, several ways have been proposed for combining
them to create a new hybrid system.
Basics of Content-based
Recommender Systems
Systems implementing a content-based recommendation approach
analyze a set of documents and/or descriptions of items previously
rated by a user, and build a model or profile of user interests based on
the features of the objects rated by that user.
The profile is a structured representation of user interests, adopted to
recommend new interesting items. The recommendation process
basically consists in matching up the attributes of the user profile
against the attributes of a content object.
The result is a relevance judgment that represents the user’s level of
interest in that object. If a profile accurately reflects user preferences,
it is of tremendous advantage for the effectiveness of an information
access process.
For instance, it could be used to filter search results by deciding
whether a user is interested in a specific Web page or not and, in the
negative case, preventing it from being displayed.
A High Level Architecture of
Content-based Systems

Content-based Information Filtering (IF) systems need proper

techniques for representing the items and producing the user profile,
and some strategies for comparing the user profile with the item
representation.
The high level architecture of a content based recommender system is
depicted in Figure.1. The recommendation process is performed in
three steps, each of which is handled by a separate component:
CONTENT ANALYZER – When information has no structure (e.g. text),
some kind of pre-processing step is needed to extract structured
relevant information. The main responsibility of the component is to
represent the content of items (e.g. documents, Web pages, news,
product descriptions, etc.) coming from information sources in a form
suitable for the next processing steps. Data items are analyzed by
feature extraction techniques in order to shift item representation from
the original information space to the target one (e.g. Web pages
represented as keyword vectors). This representation is the input to
the PROFILE LEARNER and FILTERING COMPONENT.

• PROFILE LEARNER – This module collects data representative of the

user preferences and tries to generalize this data, in order to construct
the user profile. Usually, the generalization strategy is realized through
machine learning techniques, which are able to infer a model of user
interests starting from items liked or disliked in the past
 A High Level Architecture of
Content-based Systems
. For instance, the PROFILE LEARNER of a Web page recommender can implement a
relevance feedback method in which the learning technique combines vectors of
positive and negative examples into a prototype vector representing the user profile.
Training examples are Web pages on which a positive or negative feedback has been
provided by the user;

FILTERING COMPONENT – This module exploits the user profile to suggest relevant
items by matching the profile representation against that of items to be
recommended. The result is a binary or continuous relevance judgment (computed
using some similarity metrics, the latter case resulting in a ranked list of potentially
interesting items. In the above mentioned example, the matching is realized by
computing the cosine similarity between the prototype vector and the item vectors.
The first step of the recommendation process is the one performed by the CONTENT
ANALYZER, that usually borrows techniques from Information Retrieval systems. Item
descriptions coming from Information Source are processed by the CONTENT
ANALYZER, that extracts features (keywords, n-grams, concepts, . . . ) from
unstructured text to produce a structured item representation, stored in the
repository Represented Items.
A High Level Architecture of
Content-based Systems
In order to construct and update the profile of the active user ua (user
for which recommendations must be provided) her reactions to items
are collected in some way and recorded in the repository Feedback.
These reactions, called annotations or feedback, together with the
related item descriptions, are exploited during the process of learning
a model useful to predict the actual relevance of newly presented
items. Users can also explicitly define their areas of interest as an
initial profile without providing any feedback.
Typically, it is possible to distinguish between two kinds of relevance
feedback: positive information (inferring features liked by the user)
and negative information (i.e., inferring features the user is not
interested in.
Two different techniques can be adopted for recording user’s
feedback. When a system requires the user to explicitly evaluate
items, this technique is usually referred to as “explicit feedback”; the
other technique, called “implicit feedback”, does not require any
active user involvement, in the sense that feedback is derived from
monitoring and analyzing user’s activities.
Explicit evaluations indicate how relevant or interesting an item is to
the user. There are three main approaches to get explicit relevance
feedback:
 A High Level Architecture of
Content-based Systems
like/dislike – items are classified as “relevant” or “not relevant” by adopting a simple
binary rating scale, such as in [12]; • ratings – a discrete numeric scale is usually
adopted to judge items. Alternatively, symbolic ratings are mapped to a numeric
scale, such as in Syskill & Webert where users have the possibility of rating a Web
page as hot, lukewarm, or cold; • text comments – Comments about a single item
are collected and presented to the users as a means of facilitating the decision-
making process, For instance, customer’s feedback at Amazon.com or eBay.com
might help users in deciding whether an item has been appreciated by the
community.
In order to build the profile of the active user ua, the training set TRa for ua must be
defined. TRa is a set of pairs ⟨Ik ,rk⟩, where rk is the rating provided by ua on the
item representation Ik . Given a set of item representation labeled with ratings, the
PROFILE LEARNER applies supervised learning algorithms to generate a predictive
model – the user profile – which is usually stored in a profile repository for later use
by the FILTERING COMPONENT.

Given a new item representation, the FILTERING COMPONENT predicts whether it is

likely to be of interest for the active user, by comparing features in the item
representation to those in the representation of user preferences (stored in the user
profile). Usually, the FILTERING COMPONENT implements some strategies to rank
potentially interesting items according to the relevance with respect to the user
profile.
 A High Level Architecture of
Content-based Systems

Top-ranked items are included in a list of recommendations La, that is presented to

ua.

User tastes usually change in time, therefore up-to-date information must be

maintained and provided to the PROFILE LEARNER in order to automatically update
the user profile. Further feedback is gathered on generated recommendations by
letting users state their satisfaction or dissatisfaction with items in La. After gathering
that feedback, the learning process is performed again on the new training set, and
the resulting profile is adapted to the updated user interests. The iteration of the
feedback-learning cycle over time allows the system to take into account the dynamic
nature of user preferences
 Advantages and Drawbacks of
Content-based Filtering

USER INDEPENDENCE - Content-based recommenders exploit solely

ratings provided by the active user to build her own profile. Instead,
collaborative filtering methods need ratings from other users in order
to find the “nearest neighbors” of the active user, i.e., users that have
similar tastes since they rated the same items similarly. Then, only the
items that are most liked by the neighbors of the active user will be
recommended;
•TRANSPARENCY - Explanations on how the recommender system
works can be provided by explicitly listing content features or
descriptions that caused an item to occur in the list of
recommendations. Those features are indicators to consult in order to
decide whether to trust a recommendation. Conversely, collaborative
systems are black boxes since the only explanation for an item
recommendation is that unknown users with similar tastes liked that
item;
•NEW ITEM - Content-based recommenders are capable of
recommending items not yet rated by any user. As a consequence,
they do not suffer from the first-rater problem, which affects
collaborative recommenders which rely solely on users’ preferences to
make recommendations. Therefore, until the new item is rated by a
substantial number of users, the system would not be able to
recommend it
 Advantages and Drawbacks of
Content-based Filtering
LIMITATIONS
LIMITED CONTENT ANALYSIS - Content-based techniques have a
natural limit in the number and type of features that are associated,
whether automatically or manually, with the objects they recommend.
Domain knowledge is often needed, e.g., for movie recommendations
the system needs to know the actors and directors, and sometimes,
domain ontologies are also needed. No content-based
recommendation system can provide suitable suggestions if the
analyzed content does not contain enough information to discriminate
items the user likes from items the user does not like.
OVER-SPECIALIZATION - Content-based recommenders have no
inherent method for finding something unexpected. The system
suggests items whose scores are high when matched against the user
profile, hence the user is going to be recommended items similar to
those already rated. This drawback is also called serendipity problem
to highlight the tendency of the content-based systems to produce
recommendations with a limited degree of novelty. To give an
example, when a user has only rated movies directed by Stanley
Kubrick, she will be recommended just that kind of movies. A “perfect”
content-based technique would rarely find anything novel, limiting the
range of applications for which it would be useful.

• NEW USER - Enough ratings have to be collected before a content-

based recommender system can really understand user preferences
and provide accurate recommendations. Therefore, when few ratings
are available, as for a new user, the system will not be able to provide
reliable recommendations.
 Collaborative filtering (CF)
Collaborative filtering (CF) methods produce user specific recommendations of items
based on patterns of ratings or usage (e.g., purchases) without need for exogenous
information about either items or users.
In order to establish recommendations, CF systems need to relate two fundamentally
different entities: items and users. There are two primary approaches to facilitate
such a comparison, which constitute the two main techniques of CF: the
neighborhood approach and latent factor models.

Neighborhood methods focus on relationships between items or, alternatively,

between users. An item-item approach models the preference of a user to an item
based on ratings of similar items by the same user.
Latent factor models, such as matrix factorization (aka, SVD), comprise an alternative
approach by transforming both items and users to the same latent factor space. The
latent space tries to explain ratings by characterizing both products and users on
factors automatically inferred from user feedback.
 Matrix Factorization models
Latent factor models approach collaborative filtering with the holistic goal to uncover
latent features that explain observed ratings; examples include pLSA, neural
networks, Latent Dirichlet Allocation, and models that are induced by factorization of
the user-item ratings matrix (also known as SVD-based models).

SVD
Matrix factorization models map both users and items to a joint latent factor space of
dimensionality f , such that user-item interactions are modeled as inner products in
that space. The latent space tries to explain ratings by characterizing both products
and users on factors automatically inferred from user feedback. For example, when
the products are movies, factors might measure obvious dimensions such as comedy
vs. drama, amount of action, or orientation to children; less well defined dimensions
such as depth of character development or “quirkiness”; or completely
uninterpretable dimensions.
Accordingly, each item i is associated with a vector qi ∈ Rf , and each user u is
associated with a vector pu ∈ Rf . For a given item i, the elements of qi measure the
extent to which the item possesses those factors, positive or negative. For a given
user u, the elements of pu measure the extent of interest the user has in items that
are high on the corresponding factors (again, these may be positive or negative).

The resulting dot product, 𝑞 𝑇𝑖 p u, captures the interaction between user u and item
i—i.e., the overall interest of the user in characteristics of the item. The final rating is
created by also adding in the aforementioned baseline predictors that depend only on
the user or item. Thus, a rating is predicted by the rule.

In order to learn the model parameters ,(we minimize the regularized

squared error
 Matrix Factorization models

The constant λ4, which controls the extent of regularization, is usually determined by
cross validation. Minimization is typically performed by either stochastic gradient
descent or alternating least squares. Alternating least squares techniques rotate
between fixing the pu’s to solve for the qi’s and fixing the qi’s to solve for the pu’s.

Notice that when one of these is taken as a constant, the optimization problem is
quadratic and can be optimally solved
The algorithm loops through all ratings in the training data. For each given rating rui,
a prediction 𝑟Ƹui is made, and the associated prediction error eui ഺ rui – 𝑟Ƹui is computed.
=
For a given training case rui, we modify the parameters by moving in the opposite
direction of the gradient, yielding

prediction error

One can expect better accuracy by dedicating separate learning rates (γ) and
regularization (λ) to each type of learned parameter. Thus, for example, it is advised
to employ distinct learning rates to user biases, item biases and the factors
themselves.

SVD++
Prediction accuracy is improved by considering also implicit feedback, which provides
an additional indication of user preferences. This is especially helpful for those users
that provided much more implicit feedback than explicit one. As explained earlier,
even in cases where independent implicit feedback is absent, one can capture a
significant signal by accounting for which items users rate, regardless of their rating
value. This led to several methods that modeled a user factor by the identity of the
items he/she has rated. Here we focus on the SVD++ method, which was shown to
offer accuracy superior to SVD.
Matrix Factorization models

To this end, a second set of item factors is added, relating each item i to a factor
vector yi ∈ Rf . Those new item factors are used to characterize users based on the
set of items that they rated. The exact model is as follows:

yj=factor vector

The set R(u) contains the items rated by user u. Now, a user u is modeled as

We use a free user-factors vector, pu, which is learnt from the given explicit ratings.
This vector is complemented by the sum

which represents the perspective of implicit feedback. Since the yj’s are centered
around zero.
the sum is normalized by |R(u)|− 1/2 , in order to stabilize its variance across the
range of observed values of |R(u)|.

Model parameters are determined by minimizing the associated regularized squared

error function through stochastic gradient descent. We loop over all known ratings in
K, computing:
 Matrix Factorization models
Several types of implicit feedback can be simultaneously introduced into the model by
using extra sets of item factors. For example, if a user u has a certain kind of implicit
preference to the items in N1 (u) (e.g., she rented them), and a different type of
implicit feedback to the items in N2 (u) (e.g., she browsed them), we could use the
model.

The relative importance of each source of implicit feedback will be automatically

learned by the algorithm by its setting of the respective values of model parameters.
 Neighborhood models
The most common approach to CF is based on neighborhood models. In those
methods, a rating is estimated using known ratings made by the same user on similar
items. Better scalability and improved accuracy make the item-item approach more
favorable in many cases.

. In addition, item-item methods are more amenable to explaining the reasoning

behind predictions. This is because users are familiar with items previously preferred
by them, but do not know those allegedly like-minded users.
In general, latent factor models offer high expressive ability to describe various
aspects of the data. Thus, they tend to provide more accurate results than
neighborhood models. However, most literature and commercial systems (e.g., those
of Amazon and TiVo) are based on the neighborhood models. The prevalence of
neighborhood models is partly due to their relative simplicity. However, there are
more important reasons for real life systems to stick with those models. First, they
naturally provide intuitive explanations of the reasoning behind recommendations,
which often enhance user experience beyond what improved accuracy may achieve.

Collaborative filtering through neighborhood-based interpolation is probably the most

popular way to create a recommender system. Three major components characterize
the neighborhood approach:

(1) data normalization,

(2) neighbor selection, and
(3) determination of interpolation weights.

Normalization is essential to collaborative filtering in general, and in particular to the

more local neighborhood methods. Otherwise, even more sophisticated methods are
bound to fail, as they mix incompatible ratings pertaining to different unnormalized
users or items.

Neighborhood selection is another important component. It is directly related to the

employed similarity measure. Here, we emphasized the importance of shrinking
unreliable similarities, in order to avoid detection of neighbors with a low rating
support.
Neighborhood models
Finally, the success of neighborhood methods depends on the choice
of the interpolation weights, which are used to estimate unknown
ratings from neighboring known ones. Nevertheless, most known
methods lack a rigorous way to derive these weights

Similarity measures
Central to most item-item approaches is a similarity measure between
items. Frequently, it is based on the Pearson correlation coefficient, ρi
j, which measures the tendency of users to rate items i and j similarly.
Since many ratings are unknown, some items may share only a
handful of common observed raters.
The empirical correlation coefficient, 𝜌෡j,î is based only on the common
user support. It is advised to work with residuals from the baseline
predictors to compensate for user- and item-specific deviations. Thus
the approximated correlation coefficient is given by

The set U(i, j) contains the users who rated both items i and j.
Because estimated correlations based on a greater user support are
more reliable, an appropriate similarity measure, denoted by si j, is a
shrunk correlation coefficient of the form
Neighborhood models

The variable ni j = |U(i, j)| denotes the number of users that rated both i and j. A
typical value for λ8 is 100. Suppose that the true ρi j are independent random
variables drawn from a normal distribution,
for known τ2 . The mean of 0 is justified if the bui account for both user and item
deviations from average. Meanwhile, suppose that

Similarity-based interpolation

Here we describe the most popular approach to neighborhood modeling, and

apparently also to CF in general. Our goal is to predict rui – the unobserved rating by
user u for item i. Using the similarity measure, we identify the k items rated by u that
are most similar to i. This set of k neighbors is denoted by Sk (i;u). The predicted
value of rui is taken as a weighted average of the ratings of neighboring items, while
adjusting for user and item effects through the baseline predictors
Neighborhood models
Note the dual use of the similarities for both identification of nearest neighbors and
as the interpolation weights in equation.
Sometimes, instead of relying directly on the similarity weights as interpolation
coefficients, one can achieve better results by transforming these weights. For
example, we have found at several datasets that squaring the correlation-based
similarities is helpful. This leads to a rule like:

Similarity-based methods became very popular because they are intuitive and
relatively simple to implement. They also offer the following two useful properties:
1.Explainability. Users expect a system to give a reason for its predictions, rather
than presenting “black box” recommendations. Explanations not only enrich the user
experience, but also encourage users to interact with the system, fix wrong
impressions and improve long-term accuracy. The neighborhood framework allows
identifying which of the past user actions are most influential on the computed
prediction.

2. New ratings. Item-item neighborhood models can provide updated

recommendations immediately after users enter new ratings. This includes handling
new users as soon as they provide feedback to the system, without needing to
retrain the model and estimate new parameters
Neighborhood models
However, standard neighborhood-based methods raise some concerns:
The similarity function (si j), which directly defines the interpolation
weights, is arbitrary
Previous neighborhood-based methods do not account for interactions
among neighbors. Each similarity between an item i and a neighbor j ∈
Sk(i;u) is computed independently of the content of Sk (i;u) and the
other similarities: sil for l ∈ Sk (i;u) − { j}.
By definition, the interpolation weights sum to one, which may cause
overfitting. Suppose that an item has no useful neighbors rated by a
particular user.
Neighborhood methods may not work well if variability of ratings
differs substantially among neighbors

Some of these issues can be fixed to a certain degree, while others are
more difficult to solve within the basic framework. For example, the
third item, dealing with the sum-to-one constraint, can be alleviated by
using the following prediction rule:
Neighborhood models

Jointly derived interpolation weights

Given a set of neighbors Sk (i;u) ,we need to compute interpolation weights

that enable the best prediction rule of the form.

Typical values of k (number of neighbors) lie in the range of 20–50. where

is the residual rating.

Formal model

To start, we consider a hypothetical dense case, where all users but u rated both i
and all its neighbors in Sk (i;u). In that case, we could learn the interpolation weights
by modeling the relationships between item i and its neighbors through a least
squares problem

Notice that the

𝑖𝑗 only unknowns here are the 𝜃 ’s. The optimal solution to the least
𝑢

squares problem is found by differentiation as a solution of a linear system of

equations. From a statistics viewpoint, it is equivalent to the result of a linear
regression (without intercept) of zvi on the zv j for j ∈ Sk (i;u). Specifically, the
optimal weights are given by
Aw = b
Neighborhood models

For a sparse ratings matrix

However, we can still estimate A and b, up to the same constant, by averaging over
the given pairwise support, leading to the following reformulation

The elements of 𝐴ҧjl or 𝑏തj may differ by orders of magnitude in terms of the
number of users included in the average. Let us denote this baseline value by avg; its
precise computation is described in the next subsection. Accordingly, we define the
corresponding k ×k matrix 𝐴ҧjl and the vector 𝑏ത∈Rk :
The parameter β controls the extent of the shrinkage. A typical value would be β =
500.

Therefore, we modify so that the interpolation weights are defined as the solution of
the linear system 𝐴መw= ෡𝑏.The resulting interpolation weights are used to predict rui.
 A global neighborhood model

A neighborhood model based on global optimization has many advantages.

1.No reliance on arbitrary or heuristic item-item similarities. The new model is cast as
the solution to a global optimization problem.
2.Inherent overfitting prevention or “risk control”: the model reverts to robust
baseline predictors, unless a user entered sufficiently many relevant ratings.
3.The model can capture the totality of weak signals encompassed in all of a user’s
ratings, not needing to concentrate only on the few ratings for most similar items.

4. The model naturally allows integrating different forms of user input, such as
explicit and implicit feedback.
5.A highly scalable implementation allows linear time and space complexity, thus
facilitating both item-item and user-user implementations to scale well to very large
datasets.
6.Time drifting aspects of the data can be integrated into the model, thereby
improving its accuracy

Building the model

In order to facilitate global optimization, we would like to abandon such user-specific
weights in favor of global item-item weights independent of a specific user.

The weight from j to i is denoted by wij and will be learned from the data through
optimization. An initial sketch of the model describes each rating rui by the equation

This rule starts with the crude, yet robust, baseline predictors (bui). Then, the
estimate is adjusted by summing over all ratings by u.
 A global neighborhood model
. Let us consider the interpretation of the weights. Usually the weights in a
neighborhood model represent interpolation coefficients relating unknown ratings to
existing ones. Here, we adopt a different viewpoint, that enables a more flexible
usage of the weights. We no longer treat weights as interpolation coefficients.

Instead, we take weights as part of adjustments, or offsets, added to the baseline

predictors. This way, the weight wi j is the extent by which we increase our baseline
prediction of rui based on the observed value of ru j. For two related items i and j, we
expect wi j to be high. Thus, whenever a user u rated j higher than expected (ru j −bu j
is high), we would like to increase our estimate for u’s rating of i by adding (ru j −bu)
wi j to the baseline prediction. Likewise, our estimate will not deviate much from the
baseline by an item j that u rated just as expected (ru j −bu j is around zero), or by an
item j that is not known to be predictive on i (wi j is close to zero).
First, we can use the form of binary user input, which was found beneficial for
factorization models. Namely, analyzing which items were rated regardless of rating
value. To this end, we add another set of weights, and rewrite as

Similarly, one could employ here another set of implicit feedback, N(u)—e.g., the set
of items rented or purchased by the user—leading to the rule

For two items i and j, an implicit preference by u for j leads us to adjust our estimate
of rui by ci j, which is expected to be high if j is predictive on i.
Employing global weights, rather than user-specific interpolation coefficients,
emphasizes the influence of missing ratings. In other words, a user’s opinion is
formed not only by what he rated, but also by what he did not rate. However, here
we do not use interpolation, so we can decouple the definitions of bui and bu j.

Accordingly, a more general prediction rule would be:

 A global neighborhood model

The constant 𝑏 ෡ ui can represent predictions of rui by other

methods such as a latent factor model. Here, we suggest the
following rule that was found to work well:

The constant α controls the extent of normalization. A non-

normalized rule (α = 0), encourages greater deviations from
baseline predictions for users that provided many ratings. On
the other hand, a fully normalized rule, eliminates the effect of
number of ratings on deviations from baseline predictions.
Now, the following equation is used when predicting rui

Thus, model parameters are learned by solving the regularized

least squares problem associated with
Lecture Notes - Links to Videos

Matrix Factorization : https://www.youtube.com/watch?v=10qVo3kxAhI

Neighborhood Models :
https://www.youtube.com/watch?v=M72Ez5YvO0I

Content Based Filtering

https://www.youtube.com/watch?v=v90un9ALRzw
 Lecture Notes - e-book reference

Ricci, F, Rokach, L. Shapira, B.Kantor, ―Recommender Systems Handbook,

First Edition, 2011.

Link:

https://www.cse.iitk.ac.in/users/nsrivast/HCC/Recommender_systems

_handbook.pdf
 Lecture Notes - PPTs

https://drive.google.com/file/d/12qQ1OzwndBOUn4_OEJzzgqGS30OOL
5dn/view?usp=sharing
 Lecture Notes – Quiz
Link for Quiz

https://forms.gle/ToQSAmU6S95wtS6p6
 Lecture Notes – References

 https://www.tutorialspoint.com/human_computer_interface/informati
on_search_and_visualization.htm
 http://what-when-how.com/artificial-intelligence/artificial-
intelligence-for-information-retrieval/
 Assignments

1. Apply Matrix Factorization method to determine the user

rating for a e commerce application CO4,K3

2. Apply Neighborhood model to determine the user rating

for a e commerce application CO4,K3

3. Compare and Contrast the collaborative Filtering and

content based recommendation system. CO4,K4

4. List out the design issues in building a

simple recommender system
 Part A Q & A (with K level and CO)

1. Define recommendation system. (CO4)(K1)

Recommender Systems (RSs) are software tools and techniques providing suggestions
for items to be of use to a user

2. Examine the broad classification of Recommendation systems? . (CO4)(K1)

Content-based, Collaborative filtering, Demographic, Knowledge-based, Community-
based and Hybrid recommender systems

3. List out the functions of Recommended Systems. (CO4)(K1)

Find Some Good Items, Find all good items, Annotation in context and Recommend a
sequence and bundle

4.What are the applications of recommendation system. . (CO4)(K1)

Increase the number of items sold
Sell more diverse items
Increase the user satisfaction
Increase user fidelity

Better understand what the user wants

5.What are the different forms of rating? . (CO4)(K1)

Ratings can take on a variety of forms

Numerical ratings such as the 1-5 stars provided in the book recommender associated
with Amazon.com.
Ordinal ratings, such as “strongly agree, agree, neutral, disagree, strongly disagree”
where the user is asked to select the term that best indicates her opinion regarding an
item (usually via questionnaire).
Binary ratings that model choices in which the user is simply asked to decide if a certain
item is good or bad.
Unary ratings can indicate that a user has observed or purchased an item, or otherwise
rated the item positively. In such cases, the absence of a rating indicates that we have
no information relating the user to the item (perhaps she purchased the item somewhere
else)
 Part A Q & A (with K level and CO)

6. Define web recommendation system. (CO4)(K1)

They are a particular type of personalized Web-based applications that
provide to users personalized recommendations about content they may be interested in

7. What is Collaborative filtering:? (CO4)(K1)

This approach build a model based on user past behavior an also from similar decision
made by other users. It acts as a relationship between users and items. Similarity of
items is determined by the similarity of the ratings of those items by the users who have
rated both items

8. List the advantages of Collaborative Recommendation System. (CO4)(K1)

The advantage of collaborative recommendation system is that it does not rely on
machine based content and can be used for more complex data. A large amount of data
is collected during this process in terms of information on users’ behaviors, activities or
preferences and predicting what users will like based on their similarity to other users

9. What are two modes of gathering information in Collaborative Recommendation

System? (CO4)(K1)
Implicit mode: like asking a user to rate an item, asking a user to rank a collection of
items Explicit mode: like observing the items that a user views, keeping a record of the
items that a user purchases

10.Give Examples for Content-based filtering based Applications. (CO4)(K1)

Facebook, MySpace, LinkedIn, and other social networks use collaborative filtering to
recommend new friends, groups, and other social connections

11. What is Content-based filtering? (CO4)(K1)

This approach build a model based on event properties that is by utilizing the
characteristics of an item in order to recommend other items. Similarity of items is
determined by measuring the similarity in their properties. In this approach basically user
profile is build, for creating user profile system focus on a model of the user's preference
and a history of the user's interaction with the recommender system.

12. Give Examples for Content-based filtering. (CO4)(K1)

Rotten Tomatoes, Internet Movie Database, Jinni, and Rovi Corporation

13. What is Hybrid Recommender Systems:? (CO4)(K1)

This approach is combination of collaborative filtering and content-based filtering
Examle: Netflix Mobile Recommender Systems: This is particularly context-sensitive
recommendations which are on cloud 9 in today's world.
 Part A Q & A (with K level and CO)
14. What is Matrix Factorization? CO4)(K1)
This mathematical model helps the system split an entity into multiple smaller entries,
through an ordered rectangular array of numbers or functions, to discover the features or
information underlying the interactions between users and items.

15. Where is Matrix Factorization used? CO4)(K1)

Once an individual raises a query on a search engine, the machine deploys uses matrix
factorization to generate an output in the form of recommendations. The system uses two
approaches– content-based filtering and collaborative filtering- to make recommendations.

16.What is the difference between content based filtering and collaborative filtering?
CO4)(K2)
Content based filtering - The point of content-based filtering system is to know the content
of both user and item. Usually it constructs and then compare user-profile and item-profile
using the content of shared attribute space. For example, for a movie, you represent it with
the movie stars in it and the genres (using a binary coding for example).
For user profile, you can do the same thing based on the users likes some movie
stars/genres etc
Collaborative filtering - Collaborative algorithm uses “User Behavior” for recommending
items. They exploit behavior of other users and items in terms of transaction history, ratings,
selection and purchase information. Other users behavior and preferences over the items
are used to recommend items to the new users. In this case, features of the items are not
known.

17. What are memory based RS? CO4)(K1)

The memory-based approach uses user rating data to compute the similarity between users
or items. Typical examples of this approach are neighbourhood-based CF and item-
based/user-based top-N recommendations
 Part B Qs (with K level and CO)
1. Define Recommendation based on User Ratings using appropriate
example. (13) ( CO4) (K2)

2. i) Explain Recommender system. (4) ( CO4) (K2)

i i) Explain the techniques of Matrix Factorization (9) ( CO4) (K2)

3. Explain the different types of recommendation system. ( CO4) (K2)

i) Hybrid Recommendation System (3)
ii) Content Based Recommendation System (3)
iii) Collaborative Recommendation System (3)
iv) Knowledge Based Recommendation System (4)

4.Differentiate collaborative filtering and content based systems. (13) (

CO4) (K4)

5. Define in detail about Matrix factorization models (13) ( CO4) (K2)

6. Discuss Neighbouring model in detail (13) ( CO4) (K2)

7 i) Explain is Matrix Factorization? (4) ( CO4) (K2)

ii) Discuss the approaches of recommender system. (9) ( CO4) (K2)

8. Illustrate the advantages and disadvantages of Content based and

collaborative filtering recommendation system (13) ( CO4) (K2)
 Supportive online Certification
courses
COURSERA : DATA VISUALIZATION

https://www.coursera.org/learn/datavisualization

COURSERA : WEB OF DATA

https://www.coursera.org/learn/web-data

UDEMY : Information Retrieval and Mining Massive Data Sets

https://www.udemy.com/course/information-retrieval-and-mining-massive-data-
sets/
Real time Applications in day to day
life and to Industry
Adversarial information retrieval.

Automatic summarization. Multi-document

summarization. Compound term processing.

Cross-lingual retrieval. Document

classification. Spam filtering.

Question answering.

Product Recommendation System

Job Recommendation System

 Contents beyond the Syllabus

LIMITATIONS OF RECOMMENDATION SYSTEM

Recommendation systems are widely used in a variety of applications for

recommending products or items to the user. There are two popular methods used for
filtering the recommendations, content-based and collaborative filtering. These
methods face the issue when there is not enough data to learn the relation between
user and items. In such cases, the third type of approach is used to build
the recommendation systems named as Hybrid Recommendation System. This
approach overcomes the limitations of both content-based and collaborative filtering
methods.

A hybrid recommendation system is a special type of recommendation system which

can be considered as the combination of the content and collaborative filtering
method. Combining collaborative and content-based filtering together may help in
overcoming the shortcoming we are facing at using them separately and also can be
more effective in some cases. Hybrid recommender system approaches can be
implemented in various ways like by using content and collaborative-based methods
to generate predictions separately and then combining the prediction or we can just
add the capabilities of collaborative-based methods to a content-based approach (and
vice versa).
There are several studies that compare the performance of the conventional
approaches with the hybrid methods and say that by using the hybrid methods we
can generate more accurate recommendations.
LIMITATIONS OF
RECOMMENDATION SYSTEM

Losses used by Recommendation Systems

We can build recommendation systems in two ways using two different loss
approaches:-

Bayesian Personalised Ranking(BPR) pairwise loss – this method can be used

when the positive interaction from the user on the data is presented and we
are required to optimize the ROC AUC. in this using the pairwise loss we try
to maximize the prediction difference between positive feedback and a
randomly selected negative feedback.

Weighted Approximate-Rank Pairwise(WARP) loss: this is useful when the

positive interaction is available in the feedback and we are required to
optimize some top recommendations. Here it repeatedly samples the
negative feedback until it finds the one feedback which is violating the rank
and this procedure maximizes the rank of positive feedback.
 Assessment Schedule

Model Exam

Proposed Date : 15/11/2023

Actual Date : 15/11/2023

 Prescribed Text Books & Reference
Books
TEXT BOOKS:
1.Ricardo Baeza-Yates and Berthier Ribeiro-Neto, ―Modern Information
Retrieval: The Concepts and Technology behind Search, Second Edition,
ACM Press Books, 2011.

2.Ricci, F, Rokach, L. Shapira, B.Kantor, ―Recommender Systems

Handbook, First Edition, 2011.

REFERENCE BOOKS:
1.C. Manning, P. Raghavan, and H. Schütze, ―Introduction to
Information Retrieval, Cambridge University Press, 2008.

2. Stefan Buettcher, Charles L. A. Clarke and Gordon V. Cormack,

―Information Retrieval: Implementing and Evaluating Search Engines,
The MIT Press, 2010.
 Mini Project suggestions

Search engine for Wikipedia

Sentiment analysis for Twitter

Chemistry community analysis

Document summarization

Title recommendation

Semantic correction system

Syntactic correction system

Twitter tweets classifier

Spam mail detection

 Thank you

Disclaimer:

This document is confidential and intended solely for the educational purpose of RMK Group
of Educational Institutions. If you have received this document through email in error,
please notify the system manager. This document contains proprietary information and is
intended only to the respective group / learning community as intended. If you are not the
addressee you should not disseminate, distribute or copy through e-mail. Please notify the
sender immediately by e-mail if you have received this document by mistake and delete this
document from your system. If you are not the intended recipient you are notified that
disclosing, copying, distributing or taking any action in reliance on the contents of this
information is strictly prohibited.