A Survey on Data Retrieval Techniques in Cloud Computing

S.Balasubramaniam, V.Kavitha

A Survey on Data Retrieval Techniques in Cloud Computing

S.Balasubramaniam1, Dr.V.Kavitha2
1
Full Time Ph.D Scholar, Department of Computer Science Engineering
University College of Engineering, Nagercoil - - 629004
Anna University, Chennai, Tamil Nadu, India
e-mail : baluttn@gmail.com
2
Associate Professor, Department of Computer Science Engineering
University College of Engineering, Nagercoil - - 629004
Anna University, Chennai, Tamil Nadu, India
e-mail : kavinayav@gmail.com

Abstract
A new generation of technology is transforming the world of computing. Advances in internet based
data storage, processing and services collectively known as cloud computing have emerged to
complement the traditional model of running software and storing data on personal devices or on-
premise networks. Many familiar software programs from email and word processing to spreadsheets
are now available as cloud services. Today's cloud services are highly scalable ,Which enable
customer to pay only for the computing storage and power they need, when they need it .It offers
organizations greater choice ,agility and flexibility while also driving efficiency gains and lowering
over all IT costs. Before outsourcing the data into cloud servers, data's are encrypted using encryption
algorithms by the users. That encrypted data only outsourced to cloud environment. Based on the
user's requirement retrieving the data from cloud servers is one of the challenging issues. Retrieving
the data and processing the query over cloud server is very difficult. Many searching technique are
used for retrieving the data from cloud servers. This study concentrated on various searching
algorithm for retrieving the text data from cloud servers.

Keywords: Data Retrieval, Cloud computing, Keyword

1. Introduction
Today more and more services like documentation, storage, workflows, email and office
applications like accounting ,HR ,purchase, CRM among others are being delivered form the cloud[1].
Cloud computing has as many definition as there are squares on a chess board. Cloud computing refer
to outsourced, pay as you go, on demand, somewhere in the internet [2]. There are three generally
accepted cloud service models - Software as a Service (SaaS), Infrastructure as a Service (IaaS) and
Platform as a Service (PaaS). New service models can be derived from these three basic cloud service
delivery models and referred as a specific cloud service model [3]. Many enterprises have started
adopting cloud models for basic services such as patch management, OS installation, resource upgrade,
email, web hosting services and storage. Key characteristics of all three models include convenience,
cost effectiveness, flexibility and elasticity all made possible through sharing of resources by multiple
tenants and by rapid provisioning through self service including adding additional power and storage as
possible[4]. Today's cloud services are highly scalable ,Which enable customer to pay only for the
computing storage and power they need, when they need it .It offers organizations greater
choice ,agility and flexibility while also driving efficiency gains and lowering over all IT costs[5].
Moving from a highly secure data centre to internet based cloud model will require great emphasis on
security and privacy. Migration to cloud model has implications for all types of services and
applications to be hosted on cloud and may require major modifications or enhancements in existing
processes [1].

Data Retrieval involves extracting the wanted data from a database. The two primary forms of the
retrieved data are reports and queries. To retrieve the desired data the user present a set of criteria by a

Journal of Convergence Information Technology(JCIT) 15

Volume8, Number16, November 2013
 A Survey on Data Retrieval Techniques in Cloud Computing
S.Balasubramaniam, V.Kavitha

query. Then the Data base management system, software for managing databases, selects the demanded
data from the databases. The retrieved data may be stored in a file, printed or viewed on the screen [6].
Data retrieval involves the capability to easily select data for graphic or attribute editing, updating,
querying, analysis and display. The ability to query and retrieve data based on some user defined
criteria is a necessary feature of the data storage and retrieval subsystem [7].

2. Data Retrieval Techniques

2.1. Boolean Symmetric Searchable Encryption

This paper describes the searching of cloud data over encrypted format. Here “Boolean symmetric
searchable encryption” [8] (BSSE) technique is used for searching. Boolean symmetric searchable
encryption (BSSE) uses the Boolean expression query to perform queries which are composed of
conjunction, disjunction and negation [8] of keywords. In this method, the main process is the Gram-
Schmidt orthogonalization process. This process first encodes the keywords then used in queries and
lables and inner products to perform the searching of data. This method is fully randomized, the search
is linear and the increase of labels size implies a longer computation phase for each document.But this
is only focus on simple keyword matching [8] and used only for searching Boolean Queries.

2.2. Secure Ranked Keyword Search over Encrypted Cloud Data

This paper describes the searching of cloud data using Secure Ranked Keyword Search method. This
method is the securable searching of encrypted data [9]. In a traditional keyword search, it only fully
concentrates on the “Boolean Search”. Boolean Search is only used for searching Boolean queries [9].
In this method, ranking technique used for searching an encrypted data. The “Order Preserving
Mapping Technique” [9] is also used. This Order Preserving Mapping technique protects the sensitive
score information. This method is Highly Efficient one. But this searching leads to collision in the
network. Encrypted files are largely processed after main searching. Largest encrypted files are post
processed.

2.3. Privacy-Preserving Multi-keyword Ranked Search over Encrypted Cloud Data

This paper proposes the searching of Encrypted cloud data using Privacy-Preserving Multi-keyword
Ranked Search [10] (MRSE) method. Here the Basic concept used is co-ordinate matching. Co-
ordinate matching obtains the similarity between search query and data documents. Inner product
similarity is also used to describe the Multi-keyword Ranked Search over Encrypted Cloud Data
(MRSE). Here four modules of searching are performed over encrypted cloud data [10]. The four
modules are, Encrypt Module, Client Module, Multi-Keyword Module, and Admin Module. The
features of this method are, Multi-keyword Ranked Search, Privacy-Preserving, High Efficiency it is
Eliminate unnecessary traffic and Improve Search accuracy. Similarity measurements also easily
searched [10]. The Disadvantages of this is Single Keyword search with ranking and Boolean keyword
search with ranking are not possible, Not suitable for large scale cloud data, It provide much less
semantics and this schemes are developed as crypto primitives.

2.4. Fuzzy Keyword Search Over Encrypted Data in Cloud Computing

This paper proposes the searching of Encrypted cloud data through fuzzy keyword searching [11].
The techniques which are used to obtain the fuzzy keyword search are Wild card-Based Technique
Gram-Based technique and Symbol-Based trie-Traverse Search Scheme. Wild card-Based Technique
is used to solve the problem of edit operations at the same position key words. In this the edit operation
problem is solved by Edit Distance which includes Substitution, Deletion & Insertion. Gram-Based
technique constructs the fuzzy set depends on grams [11]. The gram of a string is a substring. This
substring act as a signature for high quality efficient approximate search. In Symbol-Based trie-
Traverse Search Scheme the main idea behind this scheme is that all trapdoors sharing a common
prefix may have common nodes. All fuzzy words in the trie are finding by a depth-first search [11].

16
 A Survey on Data Retrieval Techniques in Cloud Computing
S.Balasubramaniam, V.Kavitha

Fuzzy keyword search service that aims at accommodating various types and representation
inconsistencies in different user searching inputs. Here Edit Distance can also be implemented. This
method is highly efficient. It increases the searching effectiveness, improve the space efficiency,
Optimize time efficiency and Size of the fuzzy keyword set is controllable. But large storage is
complex, Support only Boolean keyword search and not supports the ranked search problem.

2.5. Practical Techniques for Searches on Encrypted Data

This paper shows the different techniques for searching the encrypted cloud data without any loss
of data confidentiality using pseudo random function, Pseudo random generator and Sequential scan
[12]. It is achieved by cryptographic schemes. A cryptographic scheme overcomes the problem of
searching the encrypted cloud data. This scheme uses Sequential scan [12] to provide the security for
the resulting system. It has following four schemes. The Basic scheme controlled Searching, support
for Hidden Searches and the Final Scheme. It is provably secure, efficient and practical. It supports
controlled and hidden search and query isolation. This method is Simple and fast, No space and
communication over head. It is Very flexible one. But Sequential scan is not efficient if data size is
large [12]. The scheme is too slow in searching for a large number of documents. If we search too often,
server may be able to learning some information.

2.6. Confidentiality Preserving Rank-Order search

This paper proposes the searching of Encrypted cloud data through Confidentiality Preserving
Rank-Order technique [13]. Confidentiality-preserving baseline model is created and this model
prevents an untrusted data center from learning information about the query and the document
collection. Confidentiality Preserving Rank-Order technique create a framework by using secure index,
encrypted domain search and ranked retrieval [13] for retrieving the cloud data over large collections
of document. Depending upon encrypted search queries it is securely rank order the documents and
retrieving the most relevant documents from an encrypted collections of data. This method is Highly
Efficient, High accuracy of search for wide range of applications and it Protects document/query
confidentiality against an outside intruder [13]. . But it is designed only for non-encrypted data also the
Complexity of this searching is high and Protecting communication link is little more complex.

2.7 Secure Conjunctive Keyword Search over Encrypted Data

This paper proposes the Conjunctive Keyword searching of Encrypted cloud data [14]. This method
defines a security model. The protocol which allows conjuctive keyword queries are linear
communication cost, Amortized linear communication cost and Constant cost. Among this protocol,
Amortized linear communication cost uses standard hardness assumption and constant cost uses new
hardness assumption. In the total number of documents stored on the server, the capabilities size for
conjunctive queries is linear. But the most of the communication cost between the user and the server
can be done online. Each capability consists of two parts Proto-capability and Query part. The user
must give server capabilities to determine the correct set of document by using set intersection and
meta keywords [14]. The conjunctive searches based on two schemes. Conjunctive search scheme with
constant online and communication cost .Conjunctive search scheme with constant communication
cost, here the advantage is communication cost linear and Small Capabilities. But the drawbacks are It
allow secure conjunctive search with small capabilities, Only partially solves the problem of secure
Boolean search on encrypted data and complete solution requires the ability to do disjunctive keyword
search [14] securely both across and within keyword fields.

2.8 Achieving Efficient Conjunctive Keyword Searches over Encrypted Data

17
 A Survey on Data Retrieval Techniques in Cloud Computing
S.Balasubramaniam, V.Kavitha

In this paper, the objective is to achieve an efficient conjunctive keyword searching of encrypted
data [15]. Here two schemes are used to achieve an efficient conjunctive keyword searching of cloud
data. The two schemes are namely, Shamir secret sharing and bilinear pairing [15]. The first scheme is
based on Shamir Secret Sharing and provides the most efficient search technique in this context to date.
Although the size of its trapdoors is linear in the number of documents being searched, this is shows
that this overhead remains reasonable in practice. Nonetheless, to address this limitation alternative
method based on bilinear pairingsis proposed. This bilinear pairingsis that yields constant size
trapdoors. This latter construction is not only asymptotically more efficient than previous secure
conjunctive keyword search schemes [15] in the symmetric setting, but incurs significantly less storage
overhead. Additionally, unlike most previous work, these constructions are proven secure in the
standard model.

2.9 Efficient Conjunctive Keyword Search on Encrypted Data Storage System

In the previous conjunctive keyword search scheme, but the communication and storage costs
linearly depend on the number of stored data in the database, [16] hence it is not really suitable for a
large scale database. This paper proposes an efficient conjunctive keyword search scheme over
encrypted data in aspects of communication and storage costs. Here the storage cost of a user and the
communication cost between a user and a data supplier are reduced to the constant amounts. This paper
defines security model for a conjunctive keyword search scheme [16] and prove that the proposed
scheme is secure under the Decisional Bilinear Diffie-Hellman (DBDH) assumption in the random
oracle model. The main Advantages of this method are More Efficient than the previous one and
Suitable for large scale database. But it is used only in large scale databases.

2.10 Public Key Encryption with Keyword Search

This paper proposes the Public key encryption technique (PEKS) for searching the cloud data [17].
This method is obtained through keyword. Public key encryption technique (PEKS) uses the
constructions, Decision Diffie-Hellman Assumption & trapdoor permutations [17] to search on
encrypted data. This paper shows that PEKS implies Identity Based Encryption (IBE), but the converse
is currently the main problem. These constructions for PEKS are based on recent IBE constructions.
The constructions for finding a public key encryption for search on encrypted data search the encrypted
email through the email gateway by using identity based encryption method and keyword search [17].
Two constructions for PEKS are Bilinear Diffie Hellman and Trapdoor permutation. The advantages
are Gateway should learn nothing else about the encrypted email. And it does not violate user privacy.
Disadvantages of this method are the sender of the mail needs to explicitly mention what the keywords
are; and Keywords may not be relevant to the message at all [17]. It is used for single user only so we
are going to the next method with multiuser.
2.11 Public Key Encryption with Conjunctive Keyword Search and Its Extension to a
Multi-user System
This paper proposes the public key encryption technique (PEKS) for searching the cloud data [18].
Multi-user PECK scheme which can achieve an efficient computation and communication overhead
and effectively manage the storage in a server for a number of users. The advantages are Gateway
should learn nothing else about the encrypted email [18]. And it does not violate user privacy. It is used
for multi users.

2.12 Building an Encrypted and Searchable Audit Log

In this paper rely on check pointing to secure the integrity and verifiability of the audit log [19]. The
qualities of audit logs are, it is fully employed on most server systems and it is provide invaluable
access to past activity. The server systems used here are web logs and database logs [19]. The desirable
characteristics for building an encrypted & searchable audit logs are amper Resistant Verifiable, Data
Access Control and Searchability. That is this method Can check that entries are present and have not
been altered, It is control whether the entries may be sensitive to individuals or log owner and this
method Search for log on specific criteria (e.g keyword search).The Advantages of this method are

18
 A Survey on Data Retrieval Techniques in Cloud Computing
S.Balasubramaniam, V.Kavitha

encrypt all entries with a public key and Auditor downloads all entries, then decrypts them, then
performs the search [19]. The Disadvantages are Auditor sees all entries and regardless of what search
criteria were and all entries must be transmitted from server.

2.13 Searchable symmetric encryption: improved definitions an efficient constructions

This paper describes the Searching and retrieving of the outsourced data [20] by using multi user
Searchable Symmetric Encryption (SSE). This method focuses active research, several security
definitions and constructions, which is achieved by Non-adaptive setting and Adaptive adversary [20].
This method is used because it is more efficient and security guarantees than all previous
constructions, Support multi user setting and does not require authentication. But it is not suitable for
large scale cloud data and cannot accommodate high level requirements.

2.14 Enabling Secure and Efficient Ranked Keyword Search over Outsourced Cloud
Data

This paper describes the Searching and retrieving of the encrypted cloud data through ranked
keyword search [21] by using Ranked Searchable Symmetric Encryption, Order Preserving Symmetric
Encryption and One Many Order Preserving Mapping. This method is used for the purpose of Ranked
Key Word Search, Security and Performance guarantee. In this method Minimum communication and
computation over head and it avoids network traffic and unwanted file retrieval [21]. But this method
does not support for multiple keywords and Increase the search time and cost.

2.15 Fully Secure Functional Encryption: Attribute-Based Encryption and

(Hierarchical) Inner Product Encryption

This paper describes the Searching of the Attribute-Based Encrypted and (Hierarchical) Inner
Product Encrypted cloud data [22]. This paper contains two fully secure functional encryption schemes
namely, Attribute-based encryption (ABE) scheme and Predicate encryption (PE) scheme. Attribute-
based encryption (ABE) scheme provides a functionality that can be very useful in practice Predicate
encryption (PE) scheme gives predicate encryption of inner products between the ciphertext and key
vectors [22]. The advantages are, ABE systems constructing efficient systems with strong security
guarantees and PE - Ciphertext policy can be hidden. But the structure of the ciphertext is exposed to
users who cannot reveal.

2.16 Conjunctive, subset, and range queries on encrypted data

This paper proposes the Conjunctive Keyword searching of Encrypted cloud data using subset, and
range queries [23]. It builds a public-key system that supports a rich set of query predicates. It supports
arbitrary conjunctive queries for without leaking information on individual conjuncts. This method
uses the Hidden Vector Encryption for constructing and analyzing public-key systems supporting
queries on encrypted data [23]. The advantages of this method are Public-key encryption system
supporting any predicate, Gateway should not learn which conjunct failed and it is Useful for a large
variety of searching predicates. The disadvantages are similar constructions questions for conjunctive
queries. The non conjunctive compression system only has O ( ) size ciphertext but it requires O (log
n) size ciphertext and It require new ideas to improve the size of the ciphertext.

2.17 Zerber+R: Top-k Retrieval from a Confidential Index

This paper proposes the searching and Top-k Retrieval of Encrypted cloud data from a Confidential
Index [24]. It is construct a ranking model which allows top-k retrieval from confidential outsources
inverted index without information leakage and Preserve the privacy of encrypted data. It creates the
Vector space model used for rank calculation [24]. The features of this method are It Makes system

19
 A Survey on Data Retrieval Techniques in Cloud Computing
S.Balasubramaniam, V.Kavitha

highly scalable and It Minimizes information leakage. This method Prevents information overloads by
ranking, reducing bandwidth and Protect document frequency. But Accuracy of multiterm queries can
slightly affect.

2.18 Searchable Encryption Revisited: Consistency, Properties, Relation to Anonymous

IBE, and Extensions.

This paper proposes the Public key encryption [25] with keyword search must preserve the
consistency and security of search data through the following methods Anonymous HIBE, Public key
encryption with temporary keyword search and Identity based encryption with keyword search. This
method is used because it gives high level consistency namely perfect statistical and computational,
perfectly consistent PEKS encryption scheme gives zero decryption error and statistically consistent
PEKS is the analog of an encryption scheme with negligible decryption error [25]. But if the number of
keyword is infinitive it is not consistent and HIBE scheme support only two levels of identities

2.19 Privacy preserving keyword searches on remote encrypted data

This paper describes the Secure and searches on remotely stored encrypted data based on keyword
index [26]. It support two phase retrieval namely, Dictionary available on users mobile device
Dictionary not available on users mobile device. The advantage of this method are, Independent of
encryption method chosen for remote files, No public key cryptosystem is required, Pseudo-random
functions are implemented efficiently [26] and It allows to work with different file formats The
disadvantages are, It Requires additional storage overhead, specific encryption method only used It
cannot deal with compressed data and it is not guaranteed the security of encrypted data.

2.20 Public key encryption that allows PIR queries

This paper describes the Searching of the encrypted data through PIR Queries on over encrypted
documents by using the techniques PIR with PKES, [27] PIR with address and Bloom filter technique.
Bloom Filter allows us to keep the space that is used to store extra information. The main uses of this
method are Communication efficient search, Reveal partial information [27]. And Hide all information
in semantically secure way. But disadvantage of this is Correctness and privacy will be proved based
only on a computational assumption.

2.21 Deterministic and efficiently searchable encryption

This paper describes the consequence database encryption methods [28] that permit fast search
constraint constructs called RSA-DOAEP. In this paper, the aim is to obtain a concept of efficiently-
searchable encryption schemes which permit more flexible privacy to search-time trade-offs via a
technique called bucketization [28]. The purposes of these methods are it minimizes the number of bits
transmitted over the network and it permits efficient search ability. But Length preserving schemes also
needed for securing legacy code.

2.22 Secure Indexes

This paper describes the Searching of the encrypted data through Secure Indexes [29]. This method
Search the encrypted data by using secure index through secure index technique namely IND-CKA,
IND2-CKA, Z-IDX and Trapdoor generation [29]. The advantages of this method are it support
arbitrary update, Compressible Index and short trapdoor are used and it Support variable length
keywords and simple key management. But demerits of this are False positives occur and Bloom
filters are history independent

20
 A Survey on Data Retrieval Techniques in Cloud Computing
S.Balasubramaniam, V.Kavitha

2.23 Processing Private Queries over Untrusted Data Cloud through Privacy
Homomorphism

This paper proposes a holistic and efficient solution that comprises a secure traversal framework
and an encryption scheme [30] based on privacy homomorphism. Secure index traversal frame work,
based on which secure protocols are devised for classic types of queries [30]. The purposes of this
method are Scalable to large datasets and Efficiency of the query processing is high. But it does not
support for all query types such as top- queries, skyline queries and multi-way joins.

Table 1. Comparisons of Data retrieval techniques in cloud computing

Sl. Searching
paper Process used Advantage Disadvantage
No Technique

GramSchmidt Used only for

“Boolean symmetric Linear search,
orthogonalization searching Boolean
searchable Randomized,
process,Labeling and Queries.
1. [1] encryption” Focus on simple
inner products Longer computation
(BSSE). keyword
perform searching of phase.
matching
data

Ranked Keyword Network Traffic

Ranking technique
search Highly Efficient. occurs, Large amount
2. [2],[14] and Order Preserving
of Post –Processing of
Mapping Technique.
encrypted files

Eliminate traffic.
Multi Keyword Improve search
Ranked search Co-ordinate matching, accuracy. Not suitable for large
3. [3] Inner product Privacy scale data.
similarity. preserving multi
keyword used.
Symbol-Based trie- Edit Distance can Large storage
Fuzzy keyword Traverse Search be implemented. complexities. Support
search Scheme, Gram-Based Highly efficient. only Boolean
4. [4] keyword search. not
Technique, and Wild Increase searching
card-Based Technique effectiveness. support Ranked
search problem
Provably secure
Practical Efficient and Sequential scan is not
Pseudo random
Techniques practical. Support efficient if data size is
function, Pseudo
controlled and large.
5. [5] random generator,
hidden search and Too slow for large
Sequential scan,
query isolation. number of documents.
Cryptographic scheme
Simple and fast.
Highly Efficient.
Rank order High accuracy of It is designed only for
technique Secure index, search for wide non-encrypted data.
encrypted domain range of Complexity is high.
6. [6] search & ranked applications, Protecting
retrieval. protects document communication link is
and query little more complex.
confidentiality

21
 A Survey on Data Retrieval Techniques in Cloud Computing
S.Balasubramaniam, V.Kavitha

Security model for Small Not really suitable for

Secure conjunctive conjunctive keyword Capabilities. a large scale database.
keyword search searches over Size for It Can’t do Boolean
7. [7] encrypted data. conjunctive combinations of
queries is linear. queries

Size of the trapdoor is

linear in number of
Less storage documents being
Efficient overhead, More searched.
conjunctive Shamir secret sharing Communication and
8. [8] efficient than the
keyword search & bilinear pairing Storage cost linearly
previous methods.
depend number of
stored data in
database
More Efficient
Decisional secure
Effective than the previous
under Bilinear Diffie Used only in large
conjunctive one. Suitable for
9. [9] Hellman(DBDH) scale databases.
keyword search large scale
assumption in random
database.
oracle model
Gateway should Keywords may not be
Public key Decision Diffie- learn nothing else relevant to the
encryption Hellman Assumption about encrypted message.
10. [10]
technique(PEKS) & trapdoor mail. It is used for single
with keyword permutations Doesn't violate user only.
search, user privacy.
Efficient
Public key Multi user PECK computation and
11. [11] Complexity increases.
encryption scheme communication
technique(PEKS) overhead
Using weblogs and Auditor sees all
Tamper Resistant,
DB logs entries and regardless
Encrypted and Verifiable, Data
Building an of what search criteria
12. [12] searchable audit log Access Control
encrypted and was. All entries must
and Search ability
searchable audit logs be transmitted from
server.
Efficient and
security Not suitable for large
Symmetric Non adaptive setting guarantees. scale cloud data.
13. [13] encryption and Adaptive Support multi user Cannot accommodate
adversary. setting. high level
Doesn't require requirements.
authentication.
Build a public-key Useful for a large Similar constructions
Conjunctive system that supports a variety of /questions for
keyword search rich set of query searching conjunctive queries
14. [16] with subset and predicates. predicates
range queries.

22
 A Survey on Data Retrieval Techniques in Cloud Computing
S.Balasubramaniam, V.Kavitha

Based on Keyword Independent of

index. encryption Additional storage
Two phase retrieval Method chosen overhead.
Remote encrypted available for remote files.
15. [19] data search 1. Dictionary No public key Can't deal with
available on users crypto system is compressed data.
mobile device required.
2. Dictionary not
available on users
mobile device
Communication Correctness and
PKES with PIR Searching data via efficient search. privacy will be proved
queries PIR queries. Hide all based only on a
16. [20]
PIR with address and information. computational
Bloom filter tech. assumption.
consequence Communication Length preserving
RSA-DOAEP database encryption efficient search schemes also needed
15. [21] methods and for securing legacy
Bucketization code.

Many techniques are Arbitrary update, False positive occur

available: IND- compressible and bloom filters are
17. [22] Secure Indexes CKA,IND2-CKA,Z- index, and short history dependent.
IDX and trap door trap door used
generation

Private queries over Homo morphism Scalable to large Doesn't support all
un trusted cloud secure index traversal data set. queries.
provider. frame work. Efficiency of
18. [23] Query processing
high

3. Conclusion
In this study rigorous analysis is made on Data retrieval techniques which are used to retrieving
the original data from the encrypted data on the cloud environment. Many searchable techniques have
been analysed based on single keyword, multiple keyword search, Ranking, Similarity search, Fuzzy
tolerance. The ultimate goal is to enable rich search semantics in a privacy preserving manner and
efficiently support for large scale and distributed nature of cloud data. Beyond the text data search,
many ways of representing the data are available for example graph structured data search, image data
search, multimedia data search and multi dimensional search. This study only explained various data
retrieval techniques for text data.

4. References

[1] Data protection challenges in cloud computing, An Indian perspective, Study report, Trust worthy
computing, Microsoft, December 2010.
[2] http://cloudcomputing.sys-con.com/node/612375
[3] http://csrc.nist.gov/groups/sns/cloudcomputing/cloud-def-v15.doc
[4] A guide to Data governance for privacy, confidentiality and compliance, Microsoft white paper,
part 5: moving to cloud computing, August 2010.
[5] Trust worthy computing – Privacy in the cloud, Microsoft white paper, November 2010.
[6] http://en.wikipedia.org/wiki/Data_retrieval

23
 A Survey on Data Retrieval Techniques in Cloud Computing
S.Balasubramaniam, V.Kavitha

[7] http://bgis.sanbi.org/gis-primer/page31.htm
[8] Tarik Moataz, Abdullatif Shikfa, “Boolean Symmetric Searchable Encryption”, ASIA CCS '13
Proceedings of the 8th ACM SIGSAC symposium on Information computer and communications
security, .pp. 265-276, NY, USA , 2013.
[9] Cong Wang, Ning Cao, Jin Li, Kui Ren, Wenjing Lou, “Secure Ranked Keyword Search over
Encrypted Cloud Data” Distributed Computing Systems (ICDCS), 2010 IEEE 30th International
Conference 2010.
[10] Ning Cao, Cong Wang , Li, Ming , Kui Ren, Wenjing Lou, “Privacy-Preserving Multi-keyword
Ranked Search over Encrypted Cloud Data” INFOCOM, 2011 Proceedings IEEE April 2011.
[11] Jin Li , Qian Wang ; Cong Wang , Ning Cao , Kui Ren , Wenjing Lou “Fuzzy Keyword Search
Over Encrypted Data in Cloud Computing” INFOCOM, 2010 Proceedings IEEE March 2010.
[12] Daw Xiaoding Song , Wagner, D. , Perrig A. “Practical Techniques for Searches on Encrypted
Data” Security and Privacy, 2000. S&P 2000. Proceedings. 2000 IEEE Symposium May 2000.
[13] Swaminathan A, Mao Y, Su G-M, Gou H, Varna AL, He S, M. Wu, Oard D “Confidentiality
Preserving Rank-Order search” Storage security and survivability 2007-Conference Papers
(Proceedings of the 2007 ACM workshop on Storage security and survivability) pp.7-12.
[14] Philippe Golle, Jessica Staddon, Brent Waters “Secure Conjunctive Keyword Search over
Encrypted Data” Springer Berlin Heidelberg; ACNS – 2004 pp.31-45.
[15] Lucas Ballard, Seny Kamara, Fabian Monrose “Achieving Efficient Conjunctive Keyword
Searches over Encrypted Data” ICICS-2005; Springer Berlin Heidelberg. pp. 414-426.
[16] Jin Wook Byun, Dong Hoon Lee, Jongin Lim “Efficient Conjunctive Keyword Search on
Encrypted Data Storage System” Springer Berlin Heidelberg; Third European PKI Workshop:
Theory and Practice, EuroPKI 2006, Turin, Italy, June 19-20, 2006. Proceedings pp. 184-196.
[17] Dan Boneh, Giovanni Di Crescenzo, Rafail Ostrovsky, Giuseppe Persiano “Public Key Encryption
with Keyword Search” Springer Berlin Heidelberg; EUROCRYP’04, Volume: 3027 of LNCS-
2004 pp.506-522
[18] Y. Hwang, P. Lee, “Public Key Encryption with Conjunctive Keyword Search and Its Extension to
a Multi-user System” Pairing conference 2007.
[19] Brent R. Waters, Dirk Balfanz, Glenn Durfee, D. K. Smetters “Building an Encrypted and
Searchable Audit Log”11th Annual Network and Distributed Security Symposium (NDSS '04);
2004.
[20] R. Curtmola, J. A. Garay, S. Kamara, and R. Ostrovsky, “Searchable symmetric encryption:
improved definitions an efficient constructions” ACM CCS 06[20] conference 2006.
[21] C. Wang, N. Cao, J. Li, K. Ren, and W. Lou, “Enabling Secure and Efficient Ranked Keyword
Search over Outsourced Cloud Data” journal IEEE VOL:23 ISSUE:8 2012.
[22] A. Lewko, T. Okamoto, A. Sahai, K. Takashima, and B. Waters, “Fully Secure Functional
Encryption: Attribute-Based Encryption and (Hierarchical) Inner Product Encryption”
EUROCRYPT conference 2010.
[23] D. Boneh, B. Waters, “Conjunctive, subset, and range queries on encrypted data” Conference
TCCP.N: 535–554 2007.
[24] Zerr, Serge Olmedilla, Daniel Nejdl, Wolfgang Siberski, Wolf, “Zerber+R: Top-k Retrieval from
a Confidential Index” Conference EDBT P.No: 439-449 2009.
[25] M. Abdalla, M. Bellare, D. Catalano, E. Kiltz, T. Kohno, T. Lange, J., Malone-Lee, G. Neven, P.
Paillier, H. Shi, “Searchable Encryption Revisited: Consistency, Properties, Relation to
Anonymous IBE, and Extensions” conference J. Cryptol VOL:21 ISSUE:3 P.No:350-3912008.
[26] Y.-C. Chang, M. Mitzenmacher, “Privacy preserving keyword searches on remote encrypted data”
conference 2005.
[27] D. Boneh, E. Kushilevitz, R. Ostrovsky, and W. E. S. “Public key encryption that allows PIR
queries” CRYPTO conference 2007.
[28] M. Bellare, A. Boldyreva, A.ONeill, “Deterministic and efficiently searchable encryption”
conference CRYPTO 2007.
[29] Eu-Jin Goh, “Secure Indexes” conference Cryptology eprint archive 2003.
[30] Haibo Hu, Jianliang Xu, Chushi Ren, Byron Choi, “Processing Private Queries over Untrusted
Data Cloud through Privacy Homomorphism”, IEEE 27th International conference on Data
Engineering (ICDE), 601-612, 11-16 April 2011.  

24