Co-Extracting Opinion Targets and Opinion Words from Online Reviews Based On the Word Alignment Model

ABSTRACT

Mining opinion targets and opinion words from online reviews are important tasks for fine-grained opinion mining, the key component of which involves detecting opinion relations among words. To this end, this paper proposes a novel approach based on the partially-supervised alignment model, which regards identifying opinion relations as an alignment process. Then, a graph-based co-ranking algorithm is exploited to estimate the confidence of each candidate. Finally, candidates with higher confidence are extracted as opinion targets or opinion words. Compared to previous methods based on the nearest-neighbor rules, our model captures opinion relations more precisely, especially for long-span relations. Compared to syntax-based methods, our word alignment model effectively alleviates the negative effects of parsing errors when dealing with informal online texts. In particular, compared to the traditional unsupervised alignment model, the proposed model obtains better precision because of the usage of partial supervision. In addition, when estimating candidate confidence, we penalize higher-degree vertices in our graph-based co-ranking algorithm to decrease the probability of error generation. Our experimental results on three corpora with different sizes and languages show that our approach effectively outperforms state-of-the-art methods.

AIM

The main aim of this paper is a novel approach based on the partially-supervised alignment model, which regards identifying opinion relations as an alignment process. Then, a graph-based co-ranking algorithm is exploited to estimate the confidence of each candidate. Finally, candidates with higher confidence are extracted as opinion targets or opinion words.

SCOPE

The scope of this paper is our experimental results on three corpora with different sizes and languages show that our approach effectively outperforms state-of-the-art methods

EXISTING SYSTEM

Opinion target and opinion word extraction are not new tasks in opinion mining. There is significant effort focused on these tasks. They can be divided into two categories: sentence-level extraction and corpus level extraction according to their extraction aims. In sentence-level extraction, the task of opinion target/ word extraction is to identify the opinion target mentions or opinion expressions in sentences. Thus, these tasks are usually regarded as sequence-labeling problems. Intuitively, contextual words are selected as the features to indicate opinion targets/words in sentences. Most previous approaches adopted a collective unsupervised extraction framework. As mentioned in our first section, detecting opinion relations and calculating opinion associations among words are the key component of this type of method. adopted the co-occurrence frequency of opinion targets and opinion words to indicate their opinion associations. Exploited nearest-neighbor rules to identify opinion relations among words. Next, frequent and explicit product features were extracted using a bootstrapping process. Only the use of co-occurrence information or nearest-neighbor rules to detect opinion relations among words could not obtain precise results.

 DISADVANTAGES

·      This strategy cannot obtain precise results because there exist long-span modified relations and diverse opinion expressions.

·      Some errors are extracted by an iteration, they would not be filtered out in subsequent iterations.

PROPOSED SYSTEM

In this paper, propose a method based on a monolingual word alignment model (WAM). An opinion target can find its corresponding modifier through word alignment. The WAM is more robust because it does not need to parse informal texts. In addition, the WAM can integrate several intuitive factors, such as word co-occurrence frequencies and word positions, into a unified model for indicating the opinion relations among words. Thus, we expect to obtain more precise results on opinion relation identification. A constrained EM algorithm based on hill-climbing is then performed to determine all of the alignments in sentences, where the model will be consistent with these links as much as possible. A random walk based co-ranking algorithm is then proposed to estimate each candidate’s confidence on the graph. In this process, we penalize high-degree vertices to weaken their impacts and decrease the probability of a random walk running into unrelated regions on the graph. Meanwhile, we calculate the prior knowledge of candidates for indicating some noises and incorporating them into our ranking algorithm to make collaborated operations on candidate confidence estimations.

ADVANTAGES

1. The  advantages of the word alignment model for opinion relation identification, but it also has a more precise performance because of the use of partial supervision
2. The confidence of each candidate is estimated in a global process with graph co-ranking. Intuitively, the error propagation is effectively alleviated.

SYSTEM CONFIGURATION

HARDWARE REQUIREMENTS:-

·                Processor          -   Pentium –III

·                Speed                -    1.1 Ghz

·                RAM                 -    256 MB(min)

·                Hard Disk         -   20 GB

·                Floppy Drive    -    1.44 MB

·                Key Board                 -    Standard Windows Keyboard

·                Mouse               -    Two or Three Button Mouse

·                Monitor             -    SVGA

SOFTWARE REQUIREMENTS:-

·                Operating System      : Windows  7                                     

·                Front End                  : ASP.NET and C#

·                Database                   : MSSQL

·                Tool                           :Visual Studio

References                              

Kang Liu, Liheng Xu, Jun Zhao,“ Co-extracting Opinion Targets and Opinion Words from Online Reviews Based on the Word Alignment Model” IEEE Transactions on Knowledge and Data Engineering, Volume 27   Issue 3 July 2014.

A Dynamic Secure Group Sharing Framework in Public Cloud Computing

Abstract

With the popularity of group data sharing in public cloud computing, the privacy and security of group sharing data have become two major issues. The cloud provider cannot be treated as a trusted third party because of its semi-trust nature, and thus the traditional security models cannot be straightforwardly generalized into cloud based group sharing frameworks. In this paper, we propose a novel secure group sharing framework for public cloud, which can effectively take advantage of the Cloud Servers’ help but have no sensitive data being exposed to attackers and the cloud provider. The framework combines proxy signature, enhanced TGDH and proxy re-encryption together into a protocol. By applying the proxy signature technique, the group leader can effectively grant the privilege of group management to one or more chosen group members. The enhanced TGDH scheme enables the group to negotiate and update the group key pairs with the help of Cloud Servers, which does not require all of the group members been online all the time. By adopting proxy re-encryption, most computationally intensive operations can be delegate to Cloud Servers without disclosing any private information. Extensive security and performance analysis shows that our proposed scheme is highly efficient and satisfies the security requirements for public cloud based secure group sharing.

Aim

The main aim is to provide privacy and security of group sharing data in public cloud computing.

Scope

The scope is to combine proxy signature, enhanced TGDH (Tree-Based Group Diffie-Hellman) and proxy re-encryption together into a protocol to effectively grant the privilege of group management and to negotiate and update the group key pairs.

Existing System

­­The security of group communication applications can be ensured by group key agreement which can provide both backward secrecy and forward secrecy which are not totally the same as that defined in cloud based group sharing. These schemes can be divided into two categories: centralized and distributed all of which require all group members to be online together during the protocol implementation. Unfortunately, it’s difficult to have such “online together” guarantee in group applications in the cloud. How to make sure that such group applications in the cloud are secure and reliable remains a challenging problem. From what we know, only the work in makes a preliminary attempt, which provides a fully distributed TGDH (Tree Based Group Diffie-Hellman) based scheme. Although the scheme only requires asynchronous communication channels, it still requires the group members to participate in the process of protocol implementing and receive some others’ sent messages when members’ joining and/or leaving. Meanwhile, if a group member acting as a sponsor keeps in storing the private key of the shadow node, when he/she leaves the group, it is hard to keep backward secrecy in this scheme.

Disadvantages

·      With the popularity of group data sharing in public cloud computing, the privacy and security of group sharing data have become two major issues.

·      Existing system requires all of the group members been online all the time.

·      It still requires the group members to participate in the process of protocol implementing and receive some others’ sent messages when members’ joining and/or leaving. Meanwhile, if a group member acting as a sponsor keeps in storing the private key of the shadow node, when he/she leaves the group, it is hard to keep backward secrecy in this scheme.

Proposed System

Our work gives the extension to it to make more operability when any member online or offline at any time. In this project, based on Cloud Servers’ help, Group members can implement key synchronization when they become online in the next time.

1) The proposed scheme supports the updating of the group key pair whenever group members’ joining or leaving happens, which transfers most of the computational complexity and communication overhead to Cloud Servers without leaking the privacy.

2) Privilege of group management can be granted to any specific group member, which can be revoked at any time.

3) Enhanced on the original TGDH, with the help of Cloud Servers, the proposed scheme enables the group to negotiate and update the group key pairs even though not all of the group members are online together.

Any offline group member can launch group key synchronization when he/she becomes online again in the next time.

Advantages

·       Extensive security and performance analysis shows that our proposed scheme is highly efficient and satisfies the security requirements for public cloud based secure group sharing.

·      A novel secure group sharing framework for public cloud, which can effectively take advantage of the Cloud Servers’ help but have no sensitive data being exposed to attackers and the cloud provider.

System Architecture

SYSTEM CONFIGURATION

HARDWARE REQUIREMENTS:-

·                Processor          -   Pentium –III

·                Speed                -    1.1 Ghz

·                RAM                 -    256 MB(min)

·                Hard Disk         -   20 GB

·                Floppy Drive    -    1.44 MB

·                Key Board                 -    Standard Windows Keyboard

·                Mouse               -    Two or Three Button Mouse

·                Monitor             -    SVGA

SOFTWARE REQUIREMENTS:-

·                Operating System      : Windows  7                                     

·                Front End                  : ASP.NET and C#

·                Database                   : MSSQL

·                Tool                           :Visual Studio

References

Peilin Hong, Kaiping Xue “A DYNAMIC SECURE GROUP SHARING FRAMEWORK IN PUBLIC CLOUD COMPUTING”, IEEE Transactions on  Cloud Computing (Volume:2 ,  Issue: 4 ) October 2014.