Combining Bottom-up and Top-down Generation of Interactive Knowledge Maps for Enterprise Search

Our research project develops an intranet search engine with concept- browsing functionality, where the user is able to navigate the conceptual level in an interactive, automatically generated knowledge map. This knowledge map visualizes tacit, implicit knowledge, extracted from the intranet, as a network of semantic concepts. Inductive and deductive methods are combined; a text ana- lytics engine extracts knowledge structures from data inductively, and the en- terprise ontology provides a backbone structure to the process deductively. In addition to performing conventional keyword search, the user can browse the semantic network of concepts and associations to find documents and data rec- ords. Also, the user can expand and edit the knowledge network directly. As a vision, we propose a knowledge-management system that provides concept- browsing, based on a knowledge warehouse layer on top of a heterogeneous knowledge base with various systems interfaces. Such a concept browser will empower knowledge workers to interact with knowledge structures.

On Understanding How Developers Use the Spotter Search Tool

Analyzing how software engineers use the Integrated Development Environment (IDE) is essential to better understanding how engineers carry out their daily tasks. Spotter is a code search engine for the Pharo programming language. Since its inception, Spotter has been rapidly and broadly adopted within the Pharo community. However, little is known about how practitioners employ Spotter to search and navigate within the Pharo code base. This paper evaluates how software engineers use Spotter in practice. To achieve this, we remotely gather user actions called events. These events are then visually rendered using an adequate navigation tool chain. Sequences of events are represented using a visual alphabet. We found a number of usage patterns and identified underused Spotter features. Such findings are essential for improving Spotter.

Designer-driven 3D buildings generated using variable neighborhood search

This paper presents a mechanism to generate virtual buildings considering designer constraints and guidelines. This mechanism is implemented as a pipeline of different Variable Neighborhood Search (VNS) optimization processes in which several subproblems are tackled (1) rooms locations, (2) connectivity graph, and (3) element placement. The core VNS algorithm includes some variants to improve its performance, such as, for example constraint handling and biased operator selection. The optimization process uses a toolkit of construction primitives implemented as "smart objects" providing basic elements such as rooms, doors, staircases and other connectors. The paper also shows experimental results of the application of different designer constraints to a wide range of buildings from small houses to a large castle with several underground levels.

Optimizing search engines results using linear programming

When a query is passed to multiple search engines, each search engine returns a ranked list of documents. Researchers have demonstrated that combining results, in the form of a "metasearch engine", produces a significant improvement in coverage and search effectiveness. This paper proposes a linear programming mathematical model for optimizing the ranked list result of a given group of Web search engines for an issued query. An application with a numerical illustration shows the advantages of the proposed method. © 2011 Elsevier Ltd. All rights reserved.

Analysis and Coordination of Expert Statements in the Problems of Intellectual Information Search

* This work was financially supported by RFBF-04-01-00858.

Effectiveness of Title-Search vs. Full-Text Search in the Web

Search engines sometimes apply the search on the full text of documents or web-pages; but sometimes they can apply the search on selected parts of the documents only, e.g. their titles. Full-text search may consume a lot of computing resources and time. It may be possible to save resources by applying the search on the titles of documents only, assuming that a title of a document provides a concise representation of its content. We tested this assumption using Google search engine. We ran search queries that have been defined by users, distinguishing between two types of queries/users: queries of users who are familiar with the area of the search, and queries of users who are not familiar with the area of the search. We found that searches which use titles provide similar and sometimes even (slightly) better results compared to searches which use the full-text. These results hold for both types of queries/users. Moreover, we found an advantage in title-search when searching in unfamiliar areas because the general terms used in queries in unfamiliar areas match better with general terms which tend to be used in document titles.

A Variable Neighborhood Search Approach for Solving the Maximum Set Splitting Problem

This paper presents a Variable neighbourhood search (VNS) approach for solving the Maximum Set Splitting Problem (MSSP). The algorithm forms a system of neighborhoods based on changing the component for an increasing number of elements. An efficient local search procedure swaps the components of pairs of elements and yields a relatively short running time. Numerical experiments are performed on the instances known in the literature: minimum hitting set and Steiner triple systems. Computational results show that the proposed VNS achieves all optimal or best known solutions in short times. The experiments indicate that the VNS compares favorably with other methods previously used for solving the MSSP. ACM Computing Classification System (1998): I.2.8.

Towards next generation vertical search engines

As the Web evolves unexpectedly fast, information grows explosively. Useful resources become more and more difficult to find because of their dynamic and unstructured characteristics. A vertical search engine is designed and implemented towards a specific domain. Instead of processing the giant volume of miscellaneous information distributed in the Web, a vertical search engine targets at identifying relevant information in specific domains or topics and eventually provides users with up-to-date information, highly focused insights and actionable knowledge representation. As the mobile device gets more popular, the nature of the search is changing. So, acquiring information on a mobile device poses unique requirements on traditional search engines, which will potentially change every feature they used to have. To summarize, users are strongly expecting search engines that can satisfy their individual information needs, adapt their current situation, and present highly personalized search results. ^ In my research, the next generation vertical search engine means to utilize and enrich existing domain information to close the loop of vertical search engine's system that mutually facilitate knowledge discovering, actionable information extraction, and user interests modeling and recommendation. I investigate three problems in which domain taxonomy plays an important role, including taxonomy generation using a vertical search engine, actionable information extraction based on domain taxonomy, and the use of ensemble taxonomy to catch user's interests. As the fundamental theory, ultra-metric, dendrogram, and hierarchical clustering are intensively discussed. Methods on taxonomy generation using my research on hierarchical clustering are developed. The related vertical search engine techniques are practically used in Disaster Management Domain. Especially, three disaster information management systems are developed and represented as real use cases of my research work.^

The effects of semantic and syntactic instruction on user performance and satisfaction in search user interface design

The design of interfaces to facilitate user search has become critical for search engines, ecommercesites, and intranets. This study investigated the use of targeted instructional hints to improve search by measuring the quantitative effects of users' performance and satisfaction. The effects of syntactic, semantic and exemplar search hints on user behavior were evaluated in an empirical investigation using naturalistic scenarios. Combining the three search hint components, each with two levels of intensity, in a factorial design generated eight search engine interfaces. Eighty participants participated in the study and each completed six realistic search tasks. Results revealed that the inclusion of search hints improved user effectiveness, efficiency and confidence when using the search interfaces, but with complex interactions that require specific guidelines for search interface designers. These design guidelines will allow search designers to create more effective interfaces for a variety of searchapplications.

Towards Next Generation Vertical Search Engines

As the Web evolves unexpectedly fast, information grows explosively. Useful resources become more and more difficult to find because of their dynamic and unstructured characteristics. A vertical search engine is designed and implemented towards a specific domain. Instead of processing the giant volume of miscellaneous information distributed in the Web, a vertical search engine targets at identifying relevant information in specific domains or topics and eventually provides users with up-to-date information, highly focused insights and actionable knowledge representation. As the mobile device gets more popular, the nature of the search is changing. So, acquiring information on a mobile device poses unique requirements on traditional search engines, which will potentially change every feature they used to have. To summarize, users are strongly expecting search engines that can satisfy their individual information needs, adapt their current situation, and present highly personalized search results. In my research, the next generation vertical search engine means to utilize and enrich existing domain information to close the loop of vertical search engine's system that mutually facilitate knowledge discovering, actionable information extraction, and user interests modeling and recommendation. I investigate three problems in which domain taxonomy plays an important role, including taxonomy generation using a vertical search engine, actionable information extraction based on domain taxonomy, and the use of ensemble taxonomy to catch user's interests. As the fundamental theory, ultra-metric, dendrogram, and hierarchical clustering are intensively discussed. Methods on taxonomy generation using my research on hierarchical clustering are developed. The related vertical search engine techniques are practically used in Disaster Management Domain. Especially, three disaster information management systems are developed and represented as real use cases of my research work.

Search Based Clustering for Protecting Software with Diversified Updates

Reverse engineering is usually the stepping stone of a variety of at-tacks aiming at identifying sensitive information (keys, credentials, data, algo-rithms) or vulnerabilities and flaws for broader exploitation. Software applica-tions are usually deployed as identical binary code installed on millions of com-puters, enabling an adversary to develop a generic reverse-engineering strategy that, if working on one code instance, could be applied to crack all the other in-stances. A solution to mitigate this problem is represented by Software Diversity, which aims at creating several structurally different (but functionally equivalent) binary code versions out of the same source code, so that even if a successful attack can be elaborated for one version, it should not work on a diversified ver-sion. In this paper, we address the problem of maximizing software diversity from a search-based optimization point of view. The program to protect is subject to a catalogue of transformations to generate many candidate versions. The problem of selecting the subset of most diversified versions to be deployed is formulated as an optimisation problem, that we tackle with different search heuristics. We show the applicability of this approach on some popular Android apps.

Using interaction data for improving the offline and online evaluation of search engines

This thesis investigates how web search evaluation can be improved using historical interaction data. Modern search engines combine offline and online evaluation approaches in a sequence of steps that a tested change needs to pass through to be accepted as an improvement and subsequently deployed. We refer to such a sequence of steps as an evaluation pipeline. In this thesis, we consider the evaluation pipeline to contain three sequential steps: an offline evaluation step, an online evaluation scheduling step, and an online evaluation step. In this thesis we show that historical user interaction data can aid in improving the accuracy or efficiency of each of the steps of the web search evaluation pipeline. As a result of these improvements, the overall efficiency of the entire evaluation pipeline is increased. Firstly, we investigate how user interaction data can be used to build accurate offline evaluation methods for query auto-completion mechanisms. We propose a family of offline evaluation metrics for query auto-completion that represents the effort the user has to spend in order to submit their query. The parameters of our proposed metrics are trained against a set of user interactions recorded in the search engine’s query logs. From our experimental study, we observe that our proposed metrics are significantly more correlated with an online user satisfaction indicator than the metrics proposed in the existing literature. Hence, fewer changes will pass the offline evaluation step to be rejected after the online evaluation step. As a result, this would allow us to achieve a higher efficiency of the entire evaluation pipeline. Secondly, we state the problem of the optimised scheduling of online experiments. We tackle this problem by considering a greedy scheduler that prioritises the evaluation queue according to the predicted likelihood of success of a particular experiment. This predictor is trained on a set of online experiments, and uses a diverse set of features to represent an online experiment. Our study demonstrates that a higher number of successful experiments per unit of time can be achieved by deploying such a scheduler on the second step of the evaluation pipeline. Consequently, we argue that the efficiency of the evaluation pipeline can be increased. Next, to improve the efficiency of the online evaluation step, we propose the Generalised Team Draft interleaving framework. Generalised Team Draft considers both the interleaving policy (how often a particular combination of results is shown) and click scoring (how important each click is) as parameters in a data-driven optimisation of the interleaving sensitivity. Further, Generalised Team Draft is applicable beyond domains with a list-based representation of results, i.e. in domains with a grid-based representation, such as image search. Our study using datasets of interleaving experiments performed both in document and image search domains demonstrates that Generalised Team Draft achieves the highest sensitivity. A higher sensitivity indicates that the interleaving experiments can be deployed for a shorter period of time or use a smaller sample of users. Importantly, Generalised Team Draft optimises the interleaving parameters w.r.t. historical interaction data recorded in the interleaving experiments. Finally, we propose to apply the sequential testing methods to reduce the mean deployment time for the interleaving experiments. We adapt two sequential tests for the interleaving experimentation. We demonstrate that one can achieve a significant decrease in experiment duration by using such sequential testing methods. The highest efficiency is achieved by the sequential tests that adjust their stopping thresholds using historical interaction data recorded in diagnostic experiments. Our further experimental study demonstrates that cumulative gains in the online experimentation efficiency can be achieved by combining the interleaving sensitivity optimisation approaches, including Generalised Team Draft, and the sequential testing approaches. Overall, the central contributions of this thesis are the proposed approaches to improve the accuracy or efficiency of the steps of the evaluation pipeline: the offline evaluation frameworks for the query auto-completion, an approach for the optimised scheduling of online experiments, a general framework for the efficient online interleaving evaluation, and a sequential testing approach for the online search evaluation. The experiments in this thesis are based on massive real-life datasets obtained from Yandex, a leading commercial search engine. These experiments demonstrate the potential of the proposed approaches to improve the efficiency of the evaluation pipeline.

ADVERTISEMENT ALLOCATION AND TRUST MECHANISMS DESIGN IN SOCIAL NETWORKS

Social network sites (SNS), such as Facebook, Google+ and Twitter, have attracted hundreds of millions of users daily since their appearance. Within SNS, users connect to each other, express their identity, disseminate information and form cooperation by interacting with their connected peers. The increasing popularity and ubiquity of SNS usage and the invaluable user behaviors and connections give birth to many applications and business models. We look into several important problems within the social network ecosystem. The first one is the SNS advertisement allocation problem. The other two are related to trust mechanisms design in social network setting, including local trust inference and global trust evaluation. In SNS advertising, we study the problem of advertisement allocation from the ad platform's angle, and discuss its differences with the advertising model in the search engine setting. By leveraging the connection between social networks and hyperbolic geometry, we propose to solve the problem via approximation using hyperbolic embedding and convex optimization. A hyperbolic embedding method, \hcm, is designed for the SNS ad allocation problem, and several components are introduced to realize the optimization formulation. We show the advantages of our new approach in solving the problem compared to the baseline integer programming (IP) formulation. In studying the problem of trust mechanisms in social networks, we consider the existence of distrust (i.e. negative trust) relationships, and differentiate between the concept of local trust and global trust in social network setting. In the problem of local trust inference, we propose a 2-D trust model. Based on the model, we develop a semiring-based trust inference framework. In global trust evaluation, we consider a general setting with conflicting opinions, and propose a consensus-based approach to solve the complex problem in signed trust networks.

Sim-xl: A Powerful And User-friendly Tool For Peptide Cross-linking Analysis.

Chemical cross-linking has emerged as a powerful approach for the structural characterization of proteins and protein complexes. However, the correct identification of covalently linked (cross-linked or XL) peptides analyzed by tandem mass spectrometry is still an open challenge. Here we present SIM-XL, a software tool that can analyze data generated through commonly used cross-linkers (e.g., BS3/DSS). Our software introduces a new paradigm for search-space reduction, which ultimately accounts for its increase in speed and sensitivity. Moreover, our search engine is the first to capitalize on reporter ions for selecting tandem mass spectra derived from cross-linked peptides. It also makes available a 2D interaction map and a spectrum-annotation tool unmatched by any of its kind. We show SIM-XL to be more sensitive and faster than a competing tool when analyzing a data set obtained from the human HSP90. The software is freely available for academic use at http://patternlabforproteomics.org/sim-xl. A video demonstrating the tool is available at http://patternlabforproteomics.org/sim-xl/video. SIM-XL is the first tool to support XL data in the mzIdentML format; all data are thus available from the ProteomeXchange consortium (identifier PXD001677).

Profiling the Proteome of the Venom from the Social Wasp Polybia paulista: A Clue to Understand the Envenoming Mechanism

The study reported here is a classical bottom-up proteomic approach where proteins from wasp venom were extracted and separated by 2-DE; the individual protein spots were proteolytically digested and subsequently identified by using tandem mass spectrometry and database query with the protein search engine MASCOT. Eighty-four venom proteins belonging to 12 different molecular functions were identified. These proteins were classified into three groups; the first is constituted of typical venom proteins: antigens-5, hyaluronidases, phospholipases, heat shock proteins, metalloproteinases, metalloproteinase-desintegrin like proteins, serine proteinases, proteinase inhibitors, vascular endothelial growth factor-related protein, arginine kinases, Sol i-II and -II like proteins, alpha-glucosidase, and superoxide dismutases. The second contained proteins structurally related to the muscles that involves the venom reservoir. The third group, associated with the housekeeping of cells from venom glands, was composed of enzymes, membrane proteins of different types, and transcriptional factors. The composition of P. paulista venom permits us to hypothesize about a general envenoming mechanism based on five actions: (i) diffusion of venom through the tissues and to the blood, (ii) tissue, (iii) hemolysis, (iv) inflammation, and (v) allergy-played by antigen-5, PLA1, hyaluronidase, HSP 60, HSP 90, and arginine kinases.