Mobile Recommendation Engine
collaborative filtering and content based approach in hybrid manner then Genetic Algorithm for Enhancement of the Recommendation Engine. by this marketers also will get the unique characteristics of the product that must be created and also recommend to the user.
5. Problem Statement
• Google Example
– 10 billion web pages
– Average size of webpage= 20KB
– 10 billion*20KB = 200TB
– Disk read bandwidth = 50 MB/sec
– Time to read = 4 million seconds = 46+
days
– Even longer to do something useful
with the data
6. • The world is an over-crowded place
Problem Statement
7. • They all want to get our attention
Problem Statement
8. Problem Statement
• Mobile recommendations
To design such algorithm so that
overcome the current challenges and
problems of existing recommendation
system and get better accuracy to the
consumers as well as marketers
9. Search Engine vs Recommender System
“The Web is leaving the era of search and
entering one of discovery. What's the difference?
Search is what you do when you're looking for
something.
Discovery is when something wonderful that you
didn't know existed, or didn't know how to ask
for, finds you.” – CNN Money, “The race to
create a 'smart' Google
14. Top 10 High SAR rate phones
• Alcatel 1010 - 1.08
• BlackBerry Bold 9790 - 1.73
• BlackBerry Curve 9320 - 1.56
• HTC Desire X - 1.59
• Motorola V50 - 1.19
• Nokia 105 - 1.48
• Nokia 8810 - 1.14
• Nokia Asha 503 - 1.44
• Samsung S300 - 1.14
15. • We will consider standard value of SAR
is 0.9 to 1.4
• If its value is below and above to this
range means its means its dangerous
for health.
SAR(specific absorption rates)
16. Take The Survey
Few references are as follows:
• A Web-based personalized recommendation system for mobile phone
selection: Design, implementation, and evaluation
• Mobile Recommender Systems Francesco Ricci Faculty of Computer Science
Free University of Bolzano, Italy
• United State Patent Linden et al.
17. The Purpose of this study
• prioritize the design features
• prioritize the design aspects of cell phones
22. Regression Analysis for Proportions
When the response variable is a proportion or a binary
value (0 or 1), standard regression techniques must be
modified. STATGRAPHICS provides two important
procedures for this situation: Logistic Regression and Probit
Analysis. Both methods yield a prediction equation that is
constrained to lie between 0 and 1.
NOTE: We used R-language to find regression values for
data sets of different features.
23. Regression analysis
• It is used to model the relationship
between a response variable and one or
more predictor variables.
STATGRAPHICS provides a large number
of procedures for fitting different types
of regression models
30. Ranking Approaches:
– Collaborative filtering: “Tell me what
is popular amongst my peers”
– Content Based: “Show me more of
what I liked”
– Knowledge Based: “Tell me what fits
my needs”
– Hybrid
32. • Consider user x
• Find set N of other user whose ratings are
similar to x’s rating
• Estimate x’s rating based on rating of user in N
Collaborative filtering
33. • Pros:
– Extremely powerful and efficient
– Very relevant recommendations
– (1) The bigger the database,
– (2) the more the past behaviors, the
better the recommendations
Collaborative filtering
34. • Cons:
– Difficult to implement, resource and time-
consuming
– What about a new item that has never been
purchased?
Cannot be recommended
– What about a new customer who has never
bought anything? Cannot be compared to
other customers
no items can be recommended
Collaborative filtering
35. Item Profile :-
• Description of items
• Profile is a set of feature or set of
important words
• Convenient to think of important item
profile as a vector
Content Based Algorithm
36. • How to pick important words?????
• Usually from text mining
Content Based Algorithm
37. Boolean utility matrix
• Items are movies, only feature is
“actor”
• Suppose user x has watched 5 movies
• 2 movies featuring actor A
• 3 movies featuring actor B
• User profile=mean of item profile
38. PROBLEMS
• Cold Start Problem
• 1) occurs when new user or item enter
in the system
• Synonymy
• 1) when an item is represented with
two or more different names or
• entries having similar meanings
39. • Problem of providing recommendations when
there is not yet data available
• Item cold-start : A new item has been
added to the database (e.g., when a new
movie or book is released) but has not yet
received enough ratings to be
recommendable.
• User cold-start : A new user has joined the
system but their preferences are not yet
known
PROBLEMS
40. • Shilling Attacks
• when malicious user or competitor
enters into a system and starts giving
false ratings on some items Privacy
• Feeding personal information to the
recommender systems results in better
recommendation services but may lead
to issues of data privacy and security
PROBLEMS
41. • Limited Content Analysis and
Overspecialization
• The limited availability of content leads
to problems including overspecialization
• Grey Sheep
• occurs in pure CF systems where
opinions of a user do not match with
any group
PROBLEMS
44. Why Genetic Algorithms ?
• Exact methods or mathematical models
require lot of computational effort to solve
multi objective optimization problems.
• For real-life complex problems, not only
exact methods but also simple heuristic
techniques fail to obtain optimal/near-
optimal solutions efficiently.
45. Why Genetic Algorithms ?
Multiobjective evolutionary algorithms such
as genetic algorithms, nondominated sorting
genetic algorithm-II are suitable for searching
a true Results.
Widely-used in business, science, medical and
engineering
Optimization and Search Problems
Scheduling and Timetabling
46. Introduction to GA
• Genetic Algorithms are good at taking large,
potentially huge search spaces and navigating
them, looking for optimal combinations of
things, solutions you might not otherwise
find in a lifetime.”- Salvatore Mangano,
Computer Design, May 1995.
• The genetic algorithm (GA) is
a search heuristic that mimics the process of
natural evolution
47. GA is inspired from Nature
Natural Selection
Darwin's theory of evolution:-
only the organisms best adapted to their
environment tend to survive and transmit
their genetic characteristics in increasing
numbers to succeeding generations while those
less adapted tend to be eliminated.
48. Basic genetic algorithms
• Step 1: Represent the problem variable domain as a
chromosome of a fixed length, choose the size of a
chromosome population N, the crossover probability pc and
the mutation probability pm.
• Step 2: Define a fitness function to measure the
performance, or fitness, of an individual chromosome in the
problem domain. The fitness function establishes the basis
for selecting chromosomes that will be mated during
reproduction.
• Step 3: Randomly generate an initial population of
chromosomes of size N:
x1, x2 , . . . , xN
• Step 4: Calculate the fitness of each individual chromosome:
f (x1), f (x2), . . . , f (xN)
50. Flow Diagram For Mobile Recommender System
» Mobile Recommendation
Mobiles
Extract
Mobile
features
Database Extracting Records Analyzing GA
Display
Recommendations
51. Population Initialization
There are two primary methods to initialize a population
in a GA.
They are −
• Random Initialization − Populate the initial population
with completely random solutions.
• Heuristic initialization − Populate the initial population
using a known heuristic for the problem
52. • It has been observed that the entire population should
not be initialized using a heuristic, as it can result in the
population having similar solutions and very little diversity.
It has been experimentally observed that the random
solutions are the ones to drive the population to
optimality. Therefore, with heuristic initialization, we just
seed the population with a couple of good solutions, filling
up the rest with random solutions rather than filling the
entire population with heuristic based solutions.
• It has also been observed that heuristic initialization in
some cases, only effects the initial fitness of the
population, but in the end, it is the diversity of the
solutions which lead to optimality.
Population Initialization
53. Genetic Algorithms:
Recommender System
• The fitness function simply defined is a
function which takes a candidate solution
to the problem as input and produces as
output how “fit” our how “good” the
solution is with respect to the problem
in consideration.
55. • Suppose that the size of the chromosome population N
is 6, the crossover probability pc equals 0.7, and the
mutation probability pm equals 0.01. The fitness
function in our example is defined by
F(x, y, z, w)= 0.7 X + 0.6 Y + 0.55 Z + 0.5 W
59. Roulette wheel selection
• The most commonly used chromosome selection techniques is the
roulette wheel selection.
60. Crossover operator
• In our example, we have an initial population of 6
chromosomes. Thus, to establish the same population in
the next generation, the roulette wheel would be spin six
times.
• Once a pair of parent chromosomes is selected, the
crossover operator is applied.
• First, the crossover operator randomly chooses a crossover
point where two parent chromosomes “break”, and then
exchanges the chromosome parts after that point. As a
result, two new offspring are created.
• If a pair of chromosomes does not cross over, then the
chromosome cloning takes place, and the offspring are
created as exact copies of each parent.
62. Mutation operator
• Mutation represents a change in the gene.
• Mutation is a background operator. Its role is to provide a
guarantee that the search algorithm is not trapped on a local
optimum.
• The mutation operator flips a randomly selected gene in a
chromosome.
• The mutation probability is quite small in nature, and is kept
low for GAs, typically in the range between 0.001 and 0.01.
• We have taken one point crossover.
65. After 50 iteration we found
these results
RAM Processor Camera SM
8 2.3 16 64
8 2.7 16 128
6 2.1 20 64
8 2.2 15 64
7 2.2 18 128
8 2.1 19 128
Genetic Analysis shows that this recommendation
system is optimized both for consumers and
marketers.
81. FUTURE ENHANCEMENT
In future we will work on Genetic Algorithm on different
cross-over like 1-crossover,2-crossover, α-crossover, β-
crossover so that we will predict product for consumer as
well as marketers.
In future we will create a form so that we know the
current position of user that this particular user would
purchase an product or not. But user is closer to purchase
we recommend different EMI and various ways to save
money and as well as Increase the Company’s revenue.
82. We will Implement different regressions
techniques on the dataset so that we will
predict the user preference.
For combining previous 3-steps our ultimate
goal is to Implement the hybrid algorithm
for efficient and accurate results under
complex user environment.
FUTURE ENHANCEMENT