Variants in PSO

download Variants in PSO

of 17

Transcript of Variants in PSO

  • 7/31/2019 Variants in PSO

    1/17

    ASSOCIATION RULE MINING USING MEMETIC

    ALGORITHM

    Team : Ashwini V

    Divyamary PRangalakshmi B

    Sumithra M

    Under the Guidance of

    Dr. S. KanmaniProfessor

    Dept. of IT

    1

  • 7/31/2019 Variants in PSO

    2/17

    Objective

    To Propose an efficient methodology for mining

    ARs using Memetic Algorithm

    To compare the effectiveness of the algorithm

    with Adaptive and Hybrid PSO methodologies

    2

  • 7/31/2019 Variants in PSO

    3/17

    Extraction of interesting information or patterns from

    data in large databases is known as data mining.

    Data Mining

    3

  • 7/31/2019 Variants in PSO

    4/17

    Association rule mining provides valuable information

    in assessing significant correlations

    It studies the frequency of items occurring together in

    transactional databases

    Relationship will be in the form of a rule: IF {X}, THEN {Y}

    Parameters:

    Support Level

    Confidence Level

    ASSOCIATION RULES

    4

  • 7/31/2019 Variants in PSO

    5/17

    55

    DATASET DESCRIPTION

    Dataset Name No. of

    Instances

    No. of

    Attributes

    Attribute

    characteristics

    Lenses 24 3 Categorical

    HabermansSurvival

    306 3 Integer

    Car Evaluation 1728 6 Categorical

    Post Operative

    Patient

    90 8 Categorical,

    Integer

    Zoo 101 17 Categorical,

    Integer

  • 7/31/2019 Variants in PSO

    6/17

    EXISTING SYSTEM

    APRIORI ALGORITHM

    Generate all frequent itemsets and gets all confident

    association rules from those itemsets

    FP GROWTH TREE

    Stores information about frequent patterns in

    the database

    6

  • 7/31/2019 Variants in PSO

    7/17

    CONT..

    DYNAMIC ITEMSET COUNTING

    o It is alternate to apriori itemset generation

    o Considers only itemsets whose subsets are all frequent

    7

  • 7/31/2019 Variants in PSO

    8/17

    Most suitable in problems where multiple solutions are

    required

    GENETIC ALGORITHM is a global optimization method inspired

    by biological mechanisms such as evolution and hereditary

    Parallel implementation is easier

    EVOLUTIONARY ALGORITHM

    8

  • 7/31/2019 Variants in PSO

    9/17

    PARTICLE SWARM OPTIMIZATION

    Particle swarm optimization is a population based

    optimization Algorithm

    Aims to simulate social behaviors in nature found in

    insects, birds, fish, etc

    Each data itemset are represented as particles

    Driven by both personal and social influences

    9

  • 7/31/2019 Variants in PSO

    10/17

    Pitfalls in Traditional PSO

    The standard PSO algorithm can easily get trapped in

    the local optima when solving complex multimodal

    problems

    Becomes computationally inefficient as it depends on

    the function evaluators (FEs) required

    10

  • 7/31/2019 Variants in PSO

    11/17

    PSO

    Memetic

    APSO

    Hybrid

    (PSO+DE)

    (PSO+QC)

    PROPOSED SYSTEM

    11

  • 7/31/2019 Variants in PSO

    12/17

    12

    Memetic Particle Swarm Optimization scheme incorporates

    local search techniques in the standard Particle Swarm

    Optimization algorithm, resulting in an efficient and effective

    optimization method.

    Two local searches

    Solis and Wets local search strategy

    Shuffled Frog Leaping Algorithm based search strategy

    Memetic PSO

  • 7/31/2019 Variants in PSO

    13/17

    Adaptive PSO

    Adaptive particle swarm optimization (APSO) perform global

    search over the entire search space

    Identifies one of the following four evolutionary states:

    exploration, exploitation, convergence, and jumping outin

    each generation

    Enable the automatic control ofinertia weight, acceleration

    coefficients, and other algorithmic parameters at run time

    13

  • 7/31/2019 Variants in PSO

    14/17

    Hybrid

    (PSO+DE) DEPSO is a strategy of Dual Evolution (DES) based on the

    mechanism for sharing information.

    Improves the drawbacks easy to drop into region

    optimum and increases the performance with stableconvergence.

    (PSO+QC)

    The quantum theory of mechanics governs the

    movement of swarm particles along with an interpolationbased recombination operator.

    14

  • 7/31/2019 Variants in PSO

    15/17

    MONTH Rangalakshmi B

    Sumithra M

    Ashwini V

    Divya Mary P

    August -

    September

    Self Adaption of

    control parameters

    Memetic algorithm

    (PSO + Local search)

    October -

    November

    Hybrid model 1

    (PSO + DE)

    Hybrid model 2

    (PSO + QC)

    December Comparison of results and further Enhancement

    WORK PLAN

    15

  • 7/31/2019 Variants in PSO

    16/17

    BASE PAPERS

    Jiuzhong Zhang , XuemingDing.: A Multi-Swarm Self-Adaptive

    and Cooperative Particle Swarm Optimization, Engineering

    Applications of Artificial Intelligence, Volume 24,pp. 958967,

    2011.

    Zhan, Z-H. and Zhang, J. and Li, Y. and Chung, H.S-H. (2009)

    Adaptive particle swarm optimization. IEEE Transactions on

    Systems Man, and Cybernetics Part B: Cybernetics, 39 (6).

    pp. 1362-1381.

    16

  • 7/31/2019 Variants in PSO

    17/17

    REFERENCES

    Nickabadi,A. Ebadzadeh, M M. and Safabakhsh, R. A novel

    particle swarm optimization algorithm with adaptive inertia

    weight.Applied Soft Computing, 11,36583670, 2011.

    Radha Thangaraj , Millie Pant , Ajith Abraham , Pascal Bouvry

    .: Particle swarm optimization: Hybridization perspectives

    and experimental illustrations, Applied Mathematics and

    Computation 217 (2011) 52085226.

    17