I examined whether interpersonal similarity, an indicator of social closeness, influenced children’s epistemic trust in others’ testimony. Three- to 5-year-olds met two puppet informants, one of whom matched their preferences and physical attributes. Children were encouraged to request novel objects’ names from either informant, after which both informants provided conflicting labels for the unfamiliar objects. Physical and psychological commonalities with an informant differentially guided children’s learning preferences. Children subsequently heard the two informants differ in their accuracy when labeling familiar objects. For half the children the similar informant was accurate and the dissimilar informant inaccurate. Additionally, for half the children the inaccurate informant was blindfolded. Only 5-year-olds were more forgiving of the informant’s inaccuracy when blindfolded (i.e., justified), as compared to wearing a scarf (unjustified inaccuracy), and only for the dissimilar informant. These findings suggest that children’s reasoning about an informant’s state of knowledge varies with social closeness. Implications for children’s recall, mentalistic reasoning, and forgiving of mistakes are discussed. / text

As the base of the food chain, plankton affect the cycling of nutrients and organic matter within ecosystems and support production at higher trophic levels. The overall goal of this project was to examine how natural water quality fluctuations, such as changes in nutrients, temperature, and salinity, influence estuarine plankton community structure. To achieve this, I examined water quality as well as the diversity and biomass of eukaryotic plankton communities in a subtropical estuary located within the Mission-Aransas National Estuarine Research Reserve. The sampling sites included in this study consisted of three bay (Copano Bay West, Copano Bay East, Aransas Bay) and two river (Mission River Estuary, Aransas River Estuary) estuary sites. Water samples were collected monthly at the five sites from September 2011 to August 2012 and analyzed for a suite of abiotic and biotic variables. Eukaryotic plankton diversity and community structure were evaluated by using the terminal restriction fragment length polymorphism (t-RFLP) method. Although a narrow salinity gradient was present at the sampling sites, seasonal changes in water quality conditions were observed. In the river estuaries, water quality parameters defined three significant temporal periods at the Mission River Estuary site, whereas only one month differed at the Aransas River Estuary site, indicating little seasonal variation. The Copano Bay sites exhibited a seasonal pattern consisting of four periods, marked by a distinct fall (October, November, December) grouping, while Aransas Bay showed a seasonal pattern consisting of three periods, with no fall group. Even though the water quality conditions define different monthly groupings in the bay and river estuary sites, the same parameters – DOC, TDN, and pH – are the strongest drivers of the patterns at all of the sites. Seasonal and spatial distinctions in the Mission-Aransas Estuary eukaryotic plankton community composition were determined using t-RFLP. Frequent shifts in composition were apparent across samples collected at approximately bi-weekly to monthly intervals. There were significant differences (ANOSIM, p < 0.05) in community composition between the Aransas and Mission River Estuary and Aransas Bay sites. Although the overall ANOSIM tests show significance between eukaryotic plankton communities monthly and between the bay water quality periods, none of the pairwise comparisons were significantly different. However, the ANOSIM R-statistic for the monthly pairwise comparisons displays a general increasing trend over time from sampling, further highlighting the dynamic nature of the microbial eukaryotic assemblage within sites. / text

Informacijos amžiuje įvairūs tyrimo metodai generuoja milžiniškus kiekius duomenų, tačiau naudojant tik neapdorotus duomenis sunku atskleisti tiriamų objektų prigimtį, išsiaiškinti pagrindinę, kartais paslėptą struktūrą. Naudingai informacijai, o vėliau ir žinioms iš duomenų išgauti paprastai naudojami informacijos suformavimo (Information Retrieval) ir duomenų išgavimo (Data Mining) būdai. Panašumo tarp įvairių esybių nustatymas čia vaidina bene svarbiausią vaidmenį. Egzistuoja daugybė objektų tarpusavio panašumo nustatymo modelių ir teorijų, bet dauguma jų vienaip ar kitaip remiasi tam tikru objektų požymių tarpusavio palyginimu. Jau kurį laiką kalbama apie tai, kad dviejų esybių tarpusavio panašumas yra įtakojamas ne tik tų esybių požymių, bet ir konteksto, kuriame jie pasireiškia. Vadinasi dviejų esybių tarpusavio panašumas priklausomai nuo konteksto gali skirtis. Darbe aptariama kontekstinio panašumo svarba, esami konteksto apibrėžimai ir keli kontekstinį panašumą vertinantys matai. Išnagrinėjus literatūros šaltinius bus pristatytas kontekstinio panašumo formavimo būdas, kurio veikimas tiriamas naudojant dirbtinius ir realius duomenis. / Similarity is one of the most important aspects in the area of information retrieval and data mining. A new trend in similarity judgment assumes that similarity between two objects can not be expressed as a fixed value and depends on the context in which the similarity is measured. However no formal definition of context has yet been specified. In this thesis a new formal context definition is proposed and a method to measure contextual similarity is developed. The main idea behind this method is to extract context information from distinct groups of data. This method is tested with synthetic and real-world data sets comparing context similarity measure with distance based similarity. Results indicate that taking the similarity into account may cause significant changes in object similarity. This work shows the importance of context when measuring similarity.

No description available.

No description available.

No description available.

Research investigating the underlying causes and factors involved in violence and aggression has suggested there is a tendency for aggression to escalate as a means to justify prior aggression. In addition, past research has also examined the effect of perceived similarity towards the target of aggression on intensity and escalation of aggression. This study looked at the relationship between initial level of aggression and the escalation of aggression and at perceived similarity to the target of aggression as a possible factor influencing this escalation. Individuals engaging in severe initial aggression who experience higher perceived similarity to their targets of aggression should be more prone to justifying their actions and so might escalate more. To examine this, subjects could administer any of 10 levels of negative reinforcement (insults) to a learner for incorrect responses. Half of the subjects were required to practice this procedure with a mild and half with a severe insult. Results indicated that an effect of perceived similarity emerged, with individuals using less severe insults when perceived similarity to the learner was high. Contrary to predictions, high-perceived similarity to the learner stemmed escalation for participants insulting the learner with a severe insult initially. Moreover, participants who insulted with a mild insult initially escalated in their aggression when perceived similarity was high. In addition, an interaction effect of gender and perceived similarity was found, with men engaging in more severe subsequent aggression than women when perceived similarity to the target of aggression is high. The limitations, further directions, and implications of this study are discussed.

Distributed denial of service attacks (or DDoS) are a common occurrence on the internet and are becoming more intense as the bot-nets, used to launch them, grow bigger. Preventing or stopping DDoS is not possible without radically changing the internet infrastructure; various DDoS mitigation techniques have been devised with different degrees of success. All mitigation techniques share the need for a DDoS detection mechanism. DDoS detection based on traffic self-similarity estimation is a relatively new approach which is built on the notion that undis- turbed network traffic displays fractal like properties. These fractal like properties are known to degrade in presence of abnormal traffic conditions like DDoS. Detection is possible by observing the changes in the level of self-similarity in the traffic flow at the target of the attack. Existing literature assumes that DDoS traffic lacks the self-similar properties of undisturbed traffic. We show how existing bot- nets could be used to generate a self-similar traffic flow and thus break such assumptions. We then study the implications of self-similar attack traffic on DDoS detection. We find that, even when DDoS traffic is self-similar, detection is still possible. We also find that the traffic flow resulting from the superimposition of DDoS flow and legitimate traffic flow possesses a level of self-similarity that depends non-linearly on both relative traffic intensity and on the difference in self-similarity between the two incoming flows.

Image similarity or distortion assessment is fundamental to a broad range of applications throughout the field of image processing and machine vision. These include image restoration, denoising, coding, communication, interpolation, registration, fusion, classification and retrieval, as well as object detection, recognition, and tracking. Many existing image similarity measures have been proposed to work with specific types of image distortions (e.g., JPEG compression). There are also methods such as the structural similarity (SSIM) index that are applicable to a wider range of applications. However, even these "general-purpose" methods offer limited scopes in their applications. For example, SSIM does not apply or work properly when significant geometric changes exist between the two images being compared. The theory of Kolmogorov complexity provides solid groundwork for a generic information distance metric between any objects that minorizes all metrics in the class. The Normalized Information Distance (NID) metric provides a more useful framework. While appealing, the challenge lies in the implementation, mainly due to the non-computable nature of Kolmogorov complexity. To overcome this, a Normalized Compression Distance (NCD) measure was proposed, which is an effective approximation of NID and has found successful applications in the fields of bioinformatics, pattern recognition, and natural language processing. Nevertheless, the application of NID for image similarity and distortion analysis is still in its early stage. Several authors have applied the NID framework and the NCD algorithm to image clustering, image distinguishability, content-based image retrieval and video classification problems, but most reporting only moderate success. Moreover, due to their focuses on ! specific applications, the generic property of NID was not fully exploited. In this work, we aim for developing practical solutions for image distortion analysis based on the information distance framework. In particular, we propose two practical approaches to approximate NID for image similarity and distortion analysis. In the first approach, the shortest program that converts one image to another is found from a list of available transformations and a generic image similarity measure is built on computing the length of this shortest program as an approximation of the conditional Kolmogorov complexity in NID. In the second method, the complexity of the objects is approximated using Shannon entropy. Specifically we transform the reference and distorted images into wavelet domain and assume local independence among image subbands. Inspired by the Visual Information Fidelity (VIF) approach, the Gaussian Scale Mixture (GSM) model is adopted for Natural Scene Statistics (NSS) of the images to simplify the entropy computation. When applying image information distance framework in real-world applications, we find information distance measures often lead to useful features in many image processing applications. In particular, we develop a photo retouching distortion measure based on training a Gaussian kernel Support Vector Regression (SVR) model using information theoretic features extracted from a database of original and edited images. It is shown that the proposed measure is well correlated with subjective ranking of the images. Moreover, we propose a tone mapping operator parameter selection scheme for High Dynamic Range (HDR) images. The scheme attempts to find tone mapping parameters that minimize the NID of the HDR image and the resulting Low Dynamic Range (LDR) image, and thereby minimize the information loss in HDR to LDR tone mapping. The resulting images created by minimizing NID exhibit enhanced image quality.

Today, the advent of networking technologies and computer hardware have enabled more and more inexpensive PCs, various mobile devices, smart phones, PDAs, sensors and cameras to be linked to the Internet with better connectivity. In recent years, we have witnessed the emergence of several instances of distributed applications, providing infrastructures for social interactions over large-scale wide-area networks and facilitating the ways users share and publish data. User generated data today range from simple text files to (semi-) structured documents and multimedia content. With the emergence of Semantic Web, the number of features (associated with a content) that are used in order to index those large amounts of heterogenous pieces of data is growing dramatically. The feature sets associated with each content type can grow continuously as we discover new ways of describing a content in formulated terms. As the number of dimensions in the feature data grow (as high as 100 to 1000), it becomes harder and harder to search for information in a dataset due to the curse of dimensionality and it is not appropriate to use naive search methods, as their performance degrade to linear search. As an alternative, we can distribute the content and the query processing load to a set of peers in a distributed Peer-to-Peer (P2P) network and incorporate high-dimensional distributed search techniques to attack the problem. Currently, a large percentage of Internet traffic consists of video and music files shared and exchanged over P2P networks. In most present services, searching for music is performed through keyword search and naive string-matching algorithms using collaborative filtering techniques which mostly use tag based approaches. In music information retrieval (MIR) systems, the main goal is to make recommendations similar to the music that the user listens to. In these systems, techniques based on acoustic feature extraction can be employed to achieve content-based music similarity search (i.e., searching through music based on what can be heard from the music track). Using these techniques we can devise an automated measure of similarity that can replace the need for human experts (or users) who assign descriptive genre tags and meta-data to each recording and solve the famous cold-start problem associated with the collaborative filtering techniques. In this work we explore the advantages of distributed structures by efficiently distributing the content features and query processing load on the peers in a P2P network. Using a family of Locality Sensitive Hash (LSH) functions based on p-stable distributions we propose an efficient, scalable and load-balanced system, capable of performing K-Nearest-Neighbor (KNN) and Range queries. We also propose a new load-balanced indexing algorithm and evaluate it using our Java based simulator. Our results show that this P2P design ensures load-balancing and guarantees logarithmic number of hops for query processing. Our system is extensible to be used with all types of multi-dimensional feature data and it can also be employed as the main indexing scheme of a multipurpose recommendation system. / Graduate