PhD Theseshttp://drsr.daiict.ac.in//handle/123456789/342024-01-02T22:53:24Z2024-01-02T22:53:24ZDesign and Analysis of Privacy Models against Background Knowledge in Privacy - Preserving Data PublishingDesai, Nidhi Nitinhttp://drsr.daiict.ac.in//handle/123456789/9112023-02-23T04:56:39Z2021-01-01T00:00:00ZDesign and Analysis of Privacy Models against Background Knowledge in Privacy - Preserving Data Publishing
Desai, Nidhi Nitin
"Humongous amount of data gets collected by various online applications like social networks, cellular technologies, the healthcare sector, location - based services, and many more. The collected data can be accessed by third - party applications to study social and economic issues of society, leverage research, propose healthcare and business solutions, and even track a pandemic. As a result, online collected - data is a significant contributor in recent times. Despite the umpteen usefulness of online collected - data, it is vulnerable to privacy threats due to the presence of sensitive information of individual(s). Adding to that, the adversary has also become strong and powerful in terms of capabilities and access to knowledge. Knowledge is freely available in the public domain from sources like social profiles, social relations, previously published data and many more. As a result, privacy - preserving data publishing is a challenging research direction to venture upon. Our work mainly focuses on designing privacy models against background knowledge. Briefly, background knowledge is knowledge present with adversary used to disclose privacy of the individual(s). This makes background knowledge highly uncertain and inaccurate in nature as we cannot quantify the amount of knowledge present with the adversary. In this work, we design and analyze privacy solutions based on background knowledge. First of all, we propose an adversarial model against background knowledge and analyze existing and prominent privacy models against it. Secondly, we propose a privacy model (q, [lb, ub]+sp, a) - Private against background knowledge. The background knowledge assumption is comprehensive and realistic, which makes the proposed privacy model
more strong and comprehensive in nature. The proposed privacy model has been theoretically analyzed against a strong adversary. Also, the proposed privacy model has been evaluated experimentally and compared with existing literature. Progressively, our research work extends to Social Networks, which is an important application of privacy - preserving data publishing. Social network data has become an important resource in recent times but is prone to privacy threats. Thirdly, we propose a privacy model named Rule Anonymity against rule - based mining techniques in social networks. The rule - based mining techniques can predict unpublished sensitive information by generating rules. This makes it a challenging adversarial assumption. A rule - based anonymization technique has been proposed that incorporates the Rule Anonymity principle. We analyze the rule - based anonymization technique against a strong adversary having the capability of rule - based mining technique. The experimental evaluation of the rule - based anonymization technique shows positive results in terms of privacy when compared with existing literature. Fourth, we propose a de - anonymization technique against adversary’s background knowledge. The adversary’s background knowledge considers a comprehensive background knowledge that is imprecise and inaccurate in nature. We suggested distance metrics that consider imprecise and inaccurate identification and structural information. The de - anonymization technique has been implemented on a real social dataset and exhibits positive results in terms of de - anonymization accuracy. Fifth, we propose a privacy - preserving technique against comprehensive adversarial background knowledge. We have evaluated the proposed privacy model (q, [lb, ub]+sp, a) - Private on the Adult dataset and Census Income dataset and compared it with existing literature in terms of privacy. For social networks, we have used the Facebook dataset to evaluate the proposed privacy models and techniques."
2021-01-01T00:00:00ZCopy-Move Tampering: Some New Approaches of the Detection and Localization in a Digital ImageDiwan, Anjalihttp://drsr.daiict.ac.in//handle/123456789/9102023-02-23T06:30:46Z2022-01-01T00:00:00ZCopy-Move Tampering: Some New Approaches of the Detection and Localization in a Digital Image
Diwan, Anjali
Images speak. They tell us stories. Digital images carry plethora of information. The availability of cost-effective digital camera-enabled devices has made capturing images a child’s play. Statistics witnesses that the use of social networking sites has influenced people’s appetite for digital images. Billions and billions of photos are uploaded, shared and forwarded on these platforms. This makes every user an active source of the digital information. The availability of easy-to-use image editing software helps novice and experts as well capable of creating realistic alterations in these digital images. These alterations could be harmless changes for fun or serious image tampering with malicious intentions. This fact raises eyebrows and questions the authenticity of a digital image. When these digital images are used for specific purposes like news broadcasts, research publications, sports, entertainment, fashion, advertisements, legal proceedings etc., this problem becomes more critical and challenging. Therefore, digital image tampering has since long attracted the research community of image processing. Among various tampering operations, copy-move tampering is one of the easiest approaches and therefore the most common approach. In the copy-move tampering process, copying and pasting are done on the same image. Hence colour, noise component, intensity range and other properties of the image remain almost unchanged. It makes tampering detection difficult when no clue about the tampering is available other than the image itself. Further, to camouflage tampering some tricks to hide the footprint of tampering are used, such as blurring of the edges of the copy-pasted pat of the image. Technically this can be achieved by some image processing methods, e.g., JPEG compression, addition of Gaussian noise, brightness change, colour reduction, contrast adjustment etc. Occasionally some geometrical transformations such as scaling and rotation of the copied regions before pasting them somewhere else in the same image are also noticed. All these make the tampering detection a challenging task. Our study focuses on copy-move tampering detection in a digital image, either simply, i.e., without any post-processing trick or affected with different geometrical transformations and image processing methods. We first look into the tampering detection, i.e., identification and localization, using block-based technique. The first two approaches of this thesis are those, one uses LPP (Locality Preserving Projection) and the second is based on NPE (Neighborhood Preserving Embedding). Both are dimensionality reduction techniques while preserving the information of neighborhood. We find that LPP based approach worked well for simple copy-move tampering but performed poorly in case of multiple copy-move tampering and for images with self-similar structures such as some historical monuments. NPE based approach showed considerable improvement in simple copy-move images with post-processing and multiple copy-move tampering detection, however, it could not nail the tampering detection in case of self-similar images. Also, the block-based technique happens to be computationally expensive since it does a pixel-by-pixel comparison in search of a detailed clue of the tampered regions. When the copy-move region is affected with geometrical transformation, one needs a more robust clue for tampering detection. This clue must be rotation and scale-invariant. This made us concentrate on the keypoint based approach for the simple reason that image keypoints are geometrical transformation invariant. We propose to use a combination of the CenSurE keypoint detector and FREAK descriptor, which detects tampering when the image also undergoes change through scale or rotation or both following a copy-move attempt. We find that this approach also works well for simple and multiple copy-move tampering detection like in case of our two block-based approaches. The problem occurs when an image has only a few keypoints. It is observed in case of smooth images of natural landscape such as images of sky or sea or a uniform field etc. To address such situation, we propose our fourth and last approach which is based on CNN (Convolution Neural Network) and image keypoints. We have combined image information generated by CNN and CenSurE keypoints to detect and localize copy-move tampered regions. This approach enables tampering detection when the copy-move region is affected with different post-processing and geometrical transformations even in case of varying textures like smooth, coarse, or highly textured images. All these four approaches are discussed in this thesis in detail. We used several standard datasets available in public domain for performing exhaustive experiments. These are CMFD, GRIP, and CoMoFoD, MICC-F600, MICC-F220, Coverage and CASIA-II datasets. Comparison of results with some of the recently reported results of other research groups help us conclude that our approaches perform better in most of the cases and remain comparable in rest. We also discuss the future scope of our work.
2022-01-01T00:00:00ZEntity Based Query Processing For Retrieval And Summarization In Biomedical DomainSankhavara, Jainishahttp://drsr.daiict.ac.in//handle/123456789/9082023-02-23T04:58:02Z2021-01-01T00:00:00ZEntity Based Query Processing For Retrieval And Summarization In Biomedical Domain
Sankhavara, Jainisha
Exponential growth of biomedical literature poses different challenges in searching. To address complex information needs of the users, rigorous semantic processing of biomedical text is required. Biomedical information access emerges out as a new discipline for this reason. Traditional information access methods of matching, ranking, entity processing, entity-entity relationship processing, etc. are challenged in this domain. These are the major building blocks used to frame queries that represent complex information need in the area of biomedical and clinical information access. This thesis aims to do query processing using different IR and bioNLP techniques and to study their effects in retrieval and summarization. Various techniques of biomedical query reformulations are carried out and compared for biomedical document retrieval. Query expansion is one query reformulation technique which was carried out using relevance feedback and pseudo relevance feedback for biomedical document retrieval. Relevance feedback approach uses information regarding actual relevant documents to the query for feedback while pseudo relevance feedback approach does not have such information and uses top retrieved documents for feedback as they are assumed to be relevant to the query. One combined approach of relevance feedback and pseudo relevance feedback has been proposed which is based on feedback documentdiscovery and uses various classification and clustering techniques on biomedical documents to identify good document for feedback. This approach uses relevance feedback for a number of documents and tries to learn relevance for other documents for feedback. This feedback document discovery based query expansion approach shows improvement over relevance feedback based query expansion technique for biomedical document retrieval. An improved version of this feedback document discovery based query expansion approach where the features of entities are weighted based on the type of the entities and query is also proposed which shows improvement of the document retrieval system over the previous one without feature weighting. Automatic query expansion techniques based on feedback relies on two feedbacksources: feedback documents selection and feedback terms selection. In biomedical domain, medical entities are more meaningful than surface words. Therefore the entity based processing is necessary for any application in this domain. This thesis also includes a survey on advances in biomedical entity identification which includes biomedical entity identification process, various community identified challenges in the area, various resources available, approaches for biomedical entity identification and comparison of various techniques proposed in the literature for biomedical entity identification. UMLS is one biomedical resource which brings together many health and biomedical vocabularies and standards. UMLS contains biomedical entities with categorization and their relations with semantic information. A novel query expansion technique which uses knowledge from UMLS for feedback term selection is proposed where the queries are expanded using biomedical entities. The proposed method considers UMLS entities from a query with their related entities identified by UMLS and constructs query specific graph of biomedical entities for term selection. This query reformulation approach shows improvement over pseudo relevance feedback and state-of-the-art UMLS based query reformulation approaches. The amount of information for clinicians and clinical researchers is growing exponentially. These documents are long and number of topical documents are more. To synthesize the documents, text summarization attempts to reduce text so that the users can quickly understand relevant source information. In the biomedical domain, various summarization techniques are developed in recent years. Text summarization may be useful to medical practitioners with their information and knowledge management tasks. In this work we focus on query focused biomedical text summarization where the summary should be related to the query. The entity-based processing is incorporated in the summarization process along with word-embedding based similarity. The aim of this work is to use query reformulation in the summarization and see how it affects the summaries, whether expanded queries help to get better summaries.
2021-01-01T00:00:00ZDesign of Broadband Microwave Absorbers using Carbonyl Iron Filled Silicon Rubber Sheets in the Frequency Range of 1 to 8.2 GHzVashisth, Rahulhttp://drsr.daiict.ac.in//handle/123456789/9072023-02-23T05:01:01Z2022-01-01T00:00:00ZDesign of Broadband Microwave Absorbers using Carbonyl Iron Filled Silicon Rubber Sheets in the Frequency Range of 1 to 8.2 GHz
Vashisth, Rahul
For microwave absorber applications, carbonyl iron (CI) powder is a commonly used filler material in silicon rubber because of the high value of attenuation coefficients (50.56 dB/cm), high curie temperature, good temperature stabilization and the higher specific saturation magnetization intensity. Silicon rubber is preferred as the host material because of many desirable properties such as excellent weathering resistance, resistance to aging, chemical resistance, insulating properties and compatibility with many kinds of fillers. Microwave absorbers using carbonyl iron filled silicone rubber sheets (CISR) are widely used to attenuate electromagnetic interference, eliminating cavity resonances, isolating components via insertion loss, reducing harmonics and termination signals in waveguides. The fabrication accuracies in thicknesses and concentrations of CISR sheets are ± 0.5 mm and ± 2% of CI by volume, respectively. Multilayer microwave absorbers designed in this thesis have two layers because of inaccuracies in fabrication of CISR sheets. Broadband microwave absorbers are designed by using conventional type and pixelated type Frequency Selective Surface (FSS) layers embedded between two different concentrations of CISR sheets in the frequency range of 3.95 GHz to 8.2 GHz which are
light in weight and small in thickness. WR-187 (3.95 to 5.85 GHz) and WR-137 (5.85 to 8.2 GHz) rectangular dielectric waveguide (RDWG) systems are designed for non destructive testing of reflectivity, r and μr of broadband microwave absorbers. RDWG systems consist of vector network analyzer (VNA), coaxial cables, coaxial to rectangular waveguide transitions, sections of metallic waveguides, standard gain horn antennas and sections of dielectric waveguides. For 1 to 8.2 GHz, NRL arch method is used for measuring reflectivity of microwave absorbers in the anechoic chamber at normal incidence and 20° in TE/TM polarizations. This method involves measurements with two double ridge horn antennas, two long coaxial cables, wooden board for holding 1 foot × 1 foot microwave absorbers and VNA. Reflectivity of microwave abasorbers are
measured in the far field using time domain gating of VNA. Characterization ( r and μr) of CISR sheets are carried out in coaxial air line system (1 to 4 GHz) and RDWG systems (3.95 to 8.2 GHz) for six CISR sheets (0%, 10%, 20%, 30%, 40% and 50% of CI in CISR sheets by volume). In RDWG systems, flexible CISR sheets are sandwich between two Teflon sheets which are quarter wavelength at mid band. Teflon sheets provide impedance matching and avoid sagging of CISR sheets. Coaxial air line system consists of coaxial air line fixture, sections of coaxial cables and the VNA. Toroidal shaped CISR sheets are prepared and used in the coaxial air line fixture for characterization. The Nicolson- Ross-Weir (NRW) method is used to extract r and μr from measured S11 and S21 parameters of CISR sheets. For the design of microwave absorbers, r and μr value
are required for arbitrary volume fraction of CI in CISR sheets. The polynomial approximation method is used to get r and μr at intermediate concentration of CI in CISR sheets by volume. Multilayer microwave absorber in frequency range of 2.5 to 8.2 GHz is designed in this thesis using genetic optimization of concentrations and thicknesses of different layers of CISR sheets. Two layer broadband microwave absorber (7.35 mm of silicone
rubber as first layer and 1.4 mm of 50% of CI powder in CISR sheet by volume as second layer) provides reflectivity better than -10 dB in the frequency range of 2.5 to 8.2 GHz. For frequency 1.6 GHz to 2.6 GHz range, single layer microwave absorber of thickness 3 mm is designed by 50% of CI powder in CISR sheet by volume. Broadband microwave absorbers using convential type FSS and pixelated type FSS are designed fabricated and tested by embedding them between two CISR sheets (24% and 33% of CI in CISR sheets by volume). Conventional type FSS layer were designed in CST microwave studio after performing large number of simulations. Pixelated FSS layer is designed with genetic algorithm optimization and geometry refinement method. The reflectivity of both broadband FSS microwave absorbers are tested in RDWG systems and found to be better than -11 dB at normal incidence in the frequency range of 3.95 to 8.2 GHz. The two layer microwave absorber (without FSS layer) reflectivity is around -6 dB. But, by embedding FSS layer between two CISR layers increases its reflectivity to -11 dB which is low cost, light in weight. The broadband microwave absorber using pixelated FSS layer has better design methodology as it does not require human intuition, experience and a large number of simulations required in hit and trail metholodgy of conventional type FSS absorber. The analysis is performed at different incidence angle and TE / TM polarizations which shows reflectivity is generally insensitive to the incidence angles from 0° to 20°.
2022-01-01T00:00:00Z