Mubashar Mushtaq

CommuniMents: A Framework for Detecting Community Based Sentiments for Events

Social media has revolutionized human communication and styles of interaction. Due to its effectiveness and ease, people have started using it increasingly to share and exchange information, carry out discussions on various events, and express their opinions. Various communities may have diverse sentiments about events and it is an interesting research problem to understand the sentiments of a particular community for a specific event. In this article, the authors propose a framework CommuniMents which enables us to identify the members of a community and measure the sentiments of the community for a particular event. CommuniMents uses automated snowball sampling to identify the members of a community, then fetches their published contents (specifically tweets), pre-processes the contents and measures the sentiments of the community. The authors perform qualitative and quantitative evaluation for a variety of real world events to validate the effectiveness of the proposed framework.

INTWEEMS: a framework for incremental clustering of tweet streams

Twitter is a popular micro-blogging service for sharing short messages called tweets. Tweets provide public opinion on various topics. Currently twitter presents search results in form of a flat list, sorted either by popularity or by recency. These search results limit the possibility of identifying diverse latent topics covered by the tweets. One way to better understand the tweets is to cluster them where each cluster depicts a latent topic. Suitable clustering algorithms are required to cluster streaming data and map new data into existing clusters. To address this, we propose in this paper a framework called INTWEEMS (INcremental clustering of TWEEt streaMS) which clusters tweets in real-time, adjusts new tweets into existing clusters (incrementally), and provides visualization of clusters that helps in identifying latent topics and sub-topics within the tweets. This paper describes the INTWEEMS framework and its implementation.

Blind Watermarking Scheme for H.264/AVC Based on Intra 4x4 Prediction Modes

Digital video watermarking has been proposed as a scheme for copyright protection and content authentication for digital video data. H.264/AVC is a video compression standard that outperforms previous video coding standards due to its coding efficiency. In this article, we propose a blind watermarking algorithm for H.264/AVC video stream that is based on intra 4x4 prediction modes. Our proposed scheme is blind and original un-watermarked stream is not required to extract the watermark. Watermark extraction process is simple and does not require complete decoding of the video stream. Only intra 4x4 prediction modes are decoded to extract the watermark. One bit of watermark data is embedded in each of the intra 4x4 prediction modes. Mode selection based watermarking schemes are usually fragile and can be used for content authentication as well as for covert communication.

Network-aware streaming services delivery over ISP-driven P2P networks

The massive exploitation of P2P networks and the deployment of the video streaming services is bringing significant challenges for the service providers and leading to the network efficiency problem to utilize available network resources. In this paper, we propose an ISP-driven P2P framework that enables the better interaction among the ISPs and the P2P application providers. In this framework, the interaction among different entities is facilitated by a topology-aware 2-dimensional overlay network organization driven by the ISPs that is used to organize the participating peers within the network. Furthermore, this ISP-driven overlay organization is used for the streaming service delivery of the Scalable Video Coding (SVC) contents. The proposed framework intends to ensure the improved QoS for the P2P application and enables efficient utilization of the available network resources to benefit the ISPs.

Hybrid query by humming and metadata search system (HQMS)

Query by humming (QbH) is a technique that is used for audio content retrieval. Many QbH systems are based on a feature of humming comparison to audio files, which can be further improved by accompanying other approaches along with humming. In our study, we propose a Hybrid approach of QbH and Metadata search system as audio files retrieval. The proposed framework is based on the Pipe and Filter architecture that provides a serial structure with two filters in order to efficiently retrieve relevant files. Content Based searching works more swiftly when applied on a small collection of files and by using this quality our framework first filter files by audio file retrieval mechanism which will decrease the collection count to the most relevant files that would be further sieved by a second filter QbH. We find our research to be beneficial to the community, as it works on defining a new idea for audio file retrieval, made hybrid in order to achieve high precision and recall efficiently.

Enabling Cooperation between ISPs and P2P Systems toward IPTV Service Delivery

The wide adoption of the P2P networks for the deployment ofJPTV services at large scale seems very fascinating. However, it has started exposing the other side of canvas that is not acceptable for the provisioning of QoS for P2P applications. It is even more frustrating for the ISPs due to inefficient utilization of network resources. In this paper, we investigate the conflicting objectives of P2P applications & ISPs and we propose a framework enabling better interaction among entities involved in the IPTV service delivery mechanism. This interaction is facilitated by a topology-aware overlay network organization driven by the ISPs that is used to organize the participating peers within the network. The proposed ISPs-driven overlay organization facilitates the multi-attribute based sender peers selection mechanism and thus ensures the efficient utilization of available resources in ISPs domain and guarantees better QoS for the P2P applications. The framework is evaluated using the networks simulations to conform its viability. The obtained results show a noticeable improvement in the received QoS for the P2P applications and enable the efficient utilization of the available network resources to benefit the ISPs.

A fragile watermarking scheme using prediction modes for H.264/AVC content authentication

Digital watermarking has been proposed as an effective scheme for copyright protection and content authentication. Fragile video watermarking is a type of video watermarking in which embedded watermark in the digital signal is destroyed when the signal is tempered or manipulated illegally. Authenticity of video streams can be proven by embedding fragile watermarks. We propose a fragile watermarking scheme for H.264/AVC video authentication. The scheme embeds fragile watermark in intra 4×4 (I4×4) prediction modes in the intra (I) frames only during encoding. The watermark is generated from features that are robust to recompression and is embedded into the prediction modes according to defined mapping rules. Experimental results show that our proposed scheme is sensitive to recompression and has a very low effect on the video quality and bitrate. Watermark can be embedded during encoding stage and authentication is verified during decoding process.

Open issues on query by humming

Songs are considered the major source of entertainment for the people of all age groups. With the wide spread growth of World Wide Web. Various resources are available and accessible. The high availability of audio music content is bringing significant problems and the relevant songs retrieval being the foremost. Searching of audio files on the basis of its content is the most effective way, especially in the case when supportive information (metadata information) of the file is missing or incomplete. In this paper, we aim to discuss the famous content based searching technique, Query by Humming (QbH) along-with the other existing techniques in the domain. We have highlighted certain open issues and key challenges that need to be address by the research community for the advancement in the domain. The discussion is supported by conducting surveys to study the importance of these highlighted issues for the relevant songs retrial.

MFS-LDA: a multi-feature space tag recommendation model for cold start problem

Tags are used to annotate resources on social media platforms. Most tag recommendation methods use popular tags, but in the case of new resources that are as yet untagged (the cold start problem), popularity-based tag recommendation methods fail to work. The purpose of this paper is to propose a novel model for tag recommendation called multi-feature space latent Dirichlet allocation (MFS-LDA) for cold start problem.,MFS-LDA is a novel latent Dirichlet allocation (LDA)-based model which exploits multiple feature spaces (title, contents, and tags) for recommending tags. Exploiting multiple feature spaces allows MFS-LDA to recommend tags even if data from a feature space is missing (the cold start problem).,Evaluation of a publicly available data set consisting of around 20,000 Wikipedia articles that are tagged on a social bookmarking website shows a significant improvement over existing LDA-based tag recommendation methods.,The originality of MFS-LDA lies in segregation of features for removing bias toward dominant features and in synchronization of multiple feature space for tag recommendation.