Search results for: stream query

Authors: Amadou Fall Dia, Maurras Ulbricht Togbe, Aliou Boly, Zakia Kazi Aoul, Elisabeth Metais

Abstract:

Existing RDF (Resource Description Framework) Stream Processing (RSP) systems allow continuous processing of RDF data issued from different application domains such as weather station measuring phenomena, geolocation, IoT applications, drinking water distribution management, and so on. However, processing window phase often expires before finishing the entire session and RSP systems immediately delete data streams after each processed window. Such mechanism does not allow optimized exploitation of the RDF data streams as the most relevant and pertinent information of the data is often not used in a due time and almost impossible to be exploited for further analyzes. It should be better to keep the most informative part of data within streams while minimizing the memory storage space. In this work, we propose an RDF graph summarization system based on an explicit and implicit expressed needs through three main approaches: (1) an approach for user queries (SPARQL) in order to extract their needs and group them into a more global query, (2) an extension of the closeness centrality measure issued from Social Network Analysis (SNA) to determine the most informative parts of the graph and (3) an RDF graph summarization technique combining extracted user query needs and the extended centrality measure. Experiments and evaluations show efficient results in terms of memory space storage and the most expected approximate query results on summarized graphs compared to the source ones.

Keywords: centrality measures, RDF graphs summary, RDF graphs stream, SPARQL query

Procedia PDF Downloads 200

Authors: Yusuf Ziya Ayık, Ferhat Kahveci

Abstract:

Currently, database management systems have various tools such as backup and maintenance, and also provide statistical information such as resource usage and security. In terms of query performance, this paper covers query optimization, views, indexed tables, pre-computation materialized view, query performance analysis in which query plan alternatives can be created and the least costly one selected to optimize a query. Indexes and views can be created for related table columns. The literature review of this study showed that, in the course of time, despite the growing capabilities of the database management system, only database administrators are aware of the need for dealing with archival and transactional data types differently. These data may be constantly changing data used in everyday life, and also may be from the completed questionnaire whose data input was completed. For both types of data, the database uses its capabilities; but as shown in the findings section, instead of repeating similar heavy calculations which are carrying out same results with the same query over a survey results, using materialized view results can be in a more simple way. In this study, this performance difference was observed quantitatively considering the cost of the query.

Keywords: cost of query, database management systems, materialized view, query performance

Procedia PDF Downloads 277

Authors: Sana Hamdi, Emna Bouazizi, Sami Faiz

Abstract:

In recent years, real-time spatial applications, like location-aware services and traffic monitoring, have become more and more important. Such applications result dynamic environments where data as well as queries are continuously moving. As a result, there is a tremendous amount of real-time spatial data generated every day. The growth of the data volume seems to outspeed the advance of our computing infrastructure. For instance, in real-time spatial Big Data, users expect to receive the results of each query within a short time period without holding in account the load of the system. But with a huge amount of real-time spatial data generated, the system performance degrades rapidly especially in overload situations. To solve this problem, we propose the use of data partitioning as an optimization technique. Traditional horizontal and vertical partitioning can increase the performance of the system and simplify data management. But they remain insufficient for real-time spatial Big data; they can’t deal with real-time and stream queries efficiently. Thus, in this paper, we propose a novel data partitioning approach for real-time spatial Big data named VPA-RTSBD (Vertical Partitioning Approach for Real-Time Spatial Big data). This contribution is an implementation of the Matching algorithm for traditional vertical partitioning. We find, firstly, the optimal attribute sequence by the use of Matching algorithm. Then, we propose a new cost model used for database partitioning, for keeping the data amount of each partition more balanced limit and for providing a parallel execution guarantees for the most frequent queries. VPA-RTSBD aims to obtain a real-time partitioning scheme and deals with stream data. It improves the performance of query execution by maximizing the degree of parallel execution. This affects QoS (Quality Of Service) improvement in real-time spatial Big Data especially with a huge volume of stream data. The performance of our contribution is evaluated via simulation experiments. The results show that the proposed algorithm is both efficient and scalable, and that it outperforms comparable algorithms.

Keywords: real-time spatial big data, quality of service, vertical partitioning, horizontal partitioning, matching algorithm, hamming distance, stream query

Procedia PDF Downloads 156

Authors: T. G. Asmitha Chandini

Abstract:

Web-based image search re-ranking, as an successful method to get better the results. In a query keyword, the first stair is store the images is first retrieve based on the text-based information. The user to select a query keywordimage, by using this query keyword other images are re-ranked based on their visual properties with images.Now a day to day, people projected to match images in a semantic space which is used attributes or reference classes closely related to the basis of semantic image. though, understanding a worldwide visual semantic space to demonstrate highly different images from the web is difficult and inefficient. The re-ranking images, which automatically offline part learns dissimilar semantic spaces for different query keywords. The features of images are projected into their related semantic spaces to get particular images. At the online stage, images are re-ranked by compare their semantic signatures obtained the semantic précised by the query keyword image. The query-specific semantic signatures extensively improve both the proper and efficiency of image re-ranking.

Keywords: Query, keyword, image, re-ranking, semantic, signature

Procedia PDF Downloads 546

Authors: Vincent Ballola, Raymund Dilan, Thelma Palaoag

Abstract:

The Smart Online Library Catalog System with Query Expansion seeks to address the low usage of the library because of the emergence of the Internet. Library users are not accustomed to catalog systems that need a query to have the exact words without any mistakes for decent results to appear. The graphical user interface of the current system has a rather skewed learning curve for users to adapt with. With a simple graphical user interface inspired by Google, users can search quickly just by inputting their query and hitting the search button. Because of the query expansion techniques incorporated into the new system such as stemming, thesaurus search, and weighted search, users can have more efficient results from their query. The system will be adding the root words of the user's query to the query itself which will then be cross-referenced to a thesaurus database to search for any synonyms that will be added to the query. The results will then be arranged by the number of times the word has been searched. Online queries will also be added to the results for additional references. Users showed notable increases in efficiency and usability due to the familiar interface and query expansion techniques incorporated in the system. The simple yet familiar design led to a better user experience. Users also said that they would be more inclined in using the library because of the new system. The incorporation of query expansion techniques gives a notable increase of results to users that in turn gives them a wider range of resources found in the library. Used books mean more knowledge imparted to the users.

Keywords: query expansion, catalog system, stemming, weighted search, usability, thesaurus search

Procedia PDF Downloads 386

Authors: Ahmed Salisu Atiku

Abstract:

This task involves the determination of the flow arrangement for optimum performance and the calculation of total heat transfer of two identical double pipe heat exchangers in series. The inner pipe contains the cold water stream at 27°C, whilst the outer pipe contains the two hot stream of water at 50°C and 90 °C which can be mixed in any way desired. The analysis was carried out using counter flow arrangement due to its good heat transfer ability. The best way of heating this cold stream was found out to be passing the 90°C hot stream through the two heat exchangers. The outlet temperature of the cold stream was found to be 39.6°C and overall heat transfer of 131.3 kW. Though starting with 50°C hot stream in the first heat exchanger followed by 90°C hot stream in the second heat exchanger gives an outlet temperature almost the same as 90°C hot stream alone, but the heat transfer is low. The reason for the low heat transfer was that only the heat transfer in the second heat exchanger is considered. Whilst the reason behind high outlet temperature was that the cold stream was already preheated by the first stream.

Keywords: cold stream, flow arrangement, heat exchanger, hot stream

Procedia PDF Downloads 321

Authors: Dina K. Badawy, Dina M. Ibrahim, Alsayed A. Sallam

Abstract:

Distributed database is a collection of logically related databases that cooperate in a transparent manner. Query processing uses a communication network for transmitting data between sites. It refers to one of the challenges in the database world. The development of sophisticated query optimization technology is the reason for the commercial success of database systems, which complexity and cost increase with increasing number of relations in the query. Mariposa, query trading and query trading with processing task-trading strategies developed for autonomous distributed database systems, but they cause high optimization cost because of involvement of all nodes in generating an optimal plan. In this paper, we proposed a modification on the autonomous strategy K-QTPT that make the seller’s nodes with the lowest cost have gradually high priorities to reduce the optimization time. We implement our proposed strategy and present the results and analysis based on those results.

Keywords: autonomous strategies, distributed database systems, high priority, query optimization

Procedia PDF Downloads 521

Authors: Muhammet Dursun Kaya, Hasan Asil

Abstract:

In recent years, different methods have been proposed to optimize question processing in database. Although different methods have been proposed to optimize the query, but the problem which exists here is that most of these methods destroy the query execution plan after executing the query. This research attempts to solve the above problem by using a combination of methods of communicating with the database (the present questions in the programming code and using store procedures) and making query processing adaptive in database, and proposing a new approach for optimization of query processing by introducing the idea of dynamic store procedures. This research creates dynamic store procedures in the database according to the proposed algorithm. This method has been tested on applied software and results shows a significant improvement in reducing the query processing time and also reducing the workload of DBMS. Other advantages of this algorithm include: making the programming environment a single environment, eliminating the parametric limitations of the stored procedures in the database, making the stored procedures in the database dynamic, etc.

Keywords: relational database, agent, query processing, adaptable, communication with the database

Procedia PDF Downloads 369

Authors: Jae Moon Lee

Abstract:

In general, algorithms to find continuous k-nearest neighbors have been researched on the location based services, monitoring periodically the moving objects such as vehicles and mobile phone. Those researches assume the environment that the number of query points is much less than that of moving objects and the query points are not moved but fixed. In gaming environments, this problem is when computing the next movement considering the neighbors such as flocking, crowd and robot simulations. In this case, every moving object becomes a query point so that the number of query point is same to that of moving objects and the query points are also moving. In this paper, we analyze the performance of the existing algorithms focused on location based services how they operate under gaming environments.

Keywords: flocking behavior, heterogeneous agents, similarity, simulation

Procedia PDF Downloads 300

Authors: Wafa’ Alma'Aitah, Khaled Almakadmeh

Abstract:

this study presents an information retrieval system of using genetic algorithm to increase information retrieval efficiency. Using vector space model, information retrieval is based on the similarity measurement between query and documents. Documents with high similarity to query are judge more relevant to the query and should be retrieved first. Using genetic algorithms, each query is represented by a chromosome; these chromosomes are fed into genetic operator process: selection, crossover, and mutation until an optimized query chromosome is obtained for document retrieval. Results show that information retrieval with adaptive crossover probability and single point type crossover and roulette wheel as selection type give the highest recall. The proposed approach is verified using (242) proceedings abstracts collected from the Saudi Arabian national conference.

Keywords: genetic algorithm, information retrieval, optimal queries, crossover

Procedia PDF Downloads 290

Authors: Rim Ben Kalifa, Sabra Habli, Nejla Mahjoub Saïd, Hervé Bournot, Georges Le Palec

Abstract:

The present study treats different phenomena taking place in a configuration of air jet impinging on a flat surface in a coflow stream. A Computational Fluid Dynamics study is performed using the Reynolds-averaged Navier–Stokes equations by means of the Reynolds Stress Model (RSM) second order turbulent closure model. The results include mean and turbulent velocities and quantify the large effects of the coflow stream on an impinging air jet. The study of the jet in a no-directed coflow stream shows the presence of a phenomenon of recirculation near the flat plate. The influence of the coflow velocity ratio on the behavior of an impinging plane jet was also numerically investigated. The coflow stream imposed noticeable restrictions on the spreading of the impinging jet. The results show that the coflow stream decreases considerably the entrainment of air jet.

Keywords: turbulent jet, turbulence models, coflow stream, velocity ratio

Procedia PDF Downloads 237

Authors: Chun-Shien Lu

Abstract:

Privacy-preserving query has recently received considerable attention in the signal processing and multimedia community. It is also a critical step in wireless sensor network for retrieval of sensitive data. The purposes of privacy-preserving query in both the areas of signal processing and sensor network are the same, but the similarity and difference of the adopted technologies are not fully explored. In this paper, we first review the recently developed methods of privacy-preserving query, and then describe in a comprehensive manner what we can learn from the mutual of both areas.

Keywords: encryption, privacy-preserving, search, security

Procedia PDF Downloads 254

Authors: Aliyu Isah Agaie, Masrah Azrifah Azmi Murad, Nurfadhlina Mohd Sharef, Aida Mustapha

Abstract:

Natural Language Interfaces typically support a restricted language and also have scopes and limitations that naïve users are unaware of, resulting in errors when the users attempt to retrieve information from ontologies. To overcome this challenge, an auto-suggest feature is introduced into the querying process where users are guided through the querying process using interactive query construction system. Guiding users to formulate their queries, while providing them with an unconstrained (or almost unconstrained) way to query the ontology results in better interpretation of the query and ultimately lead to an effective search. The approach described in this paper is unobtrusive and subtly guides the users, so that they have a choice of either selecting from the suggestion list or typing in full. The user is not coerced into accepting system suggestions and can express himself using fragments or full sentences.

Keywords: auto-suggest, expressiveness, habitability, natural language interface, query interpretation, user guidance

Procedia PDF Downloads 472

Authors: Elaheh Vaezpour

Abstract:

Data sensing in wireless sensor networks is done by query deceleration the network by the users. In many applications of the wireless sensor networks, many users send queries to the network simultaneously. If the queries are processed separately, the network’s energy consumption will increase significantly. Therefore, it is very important to aggregate the queries before sending them to the network. In this paper, we propose a multiple query optimization framework based on sensors physical and temporal correlation. In the proposed method, queries are merged and sent to network by considering correlation among the sensors in order to reduce the communication cost between the sensors and the base station.

Keywords: wireless sensor networks, multiple query optimization, data correlation, reducing energy consumption

Procedia PDF Downloads 333

Authors: T. R. Sharika, Ashwini Kumar, Kamal Bijlani

Abstract:

Encryption is the process of converting crucial information’s unreadable to unauthorized persons. Image security is an important type of encryption that secures all type of images from cryptanalysis. A stream cipher is a fast symmetric key algorithm which is used to convert plaintext to cipher text. In this paper we are proposing an image encryption algorithm with Discrete Cosine Transform and Stream Ciphers that can improve compression of images and enhanced security. The paper also explains the use of a shuffling algorithm for enhancing securing.

Keywords: decryption, DCT, encryption, RC4 cipher, stream cipher

Procedia PDF Downloads 359

Authors: Ahmad Shahin, Fadi Chakik, Walid Moudani

Abstract:

Our study is about elaborating a potential solution for a search engine that involves semantic technology to retrieve information and display it significantly. Semantic search engines are not used widely over the web as the majorities are still in Beta stage or under construction. Many problems face the current applications in semantic search, the major problem is to analyze and calculate the meaning of query in order to retrieve relevant information. Another problem is the ontology based index and its updates. Ranking results according to concept meaning and its relation with query is another challenge. In this paper, we are offering a light meta-engine (QESM) which uses Google search, and therefore Google’s index, with some adaptations to its returned results by adding multi-query expansion. The mission was to find a reliable ranking algorithm that involves semantics and uses concepts and meanings to rank results. At the beginning, the engine finds synonyms of each query term entered by the user based on a lexical database. Then, query expansion is applied to generate different semantically analogous sentences. These are generated randomly by combining the found synonyms and the original query terms. Our model suggests the use of semantic similarity measures between two sentences. Practically, we used this method to calculate semantic similarity between each query and the description of each page’s content generated by Google. The generated sentences are sent to Google engine one by one, and ranked again all together with the adapted ranking method (QESM). Finally, our system will place Google pages with higher similarities on the top of the results. We have conducted experimentations with 6 different queries. We have observed that most ranked results with QESM were altered with Google’s original generated pages. With our experimented queries, QESM generates frequently better accuracy than Google. In some worst cases, it behaves like Google.

Keywords: semantic search engine, Google indexing, query expansion, similarity measures

Procedia PDF Downloads 424

Authors: Tim Wollert, Fabian Behrendt

Abstract:

Value Stream Management (VSM) is a widely used methodology in the context of Lean Management for improving end-to-end material and information flows from a supplier to a customer from a company’s perspective. Whereas the design principles, e.g. Pull, value-adding, customer-orientation and further ones are still valid against the background of an increasing digitalized and dynamic environment, the methodology itself for mapping a value stream is characterized as time- and resource-intensive due to the high degree of manual activities. The digitalization of processes in the context of Industry 4.0 enables new opportunities to reduce these manual efforts and make the VSM approach more agile. The paper at hand aims at providing a modular data and calculation framework, utilizing the available business data, provided by information and communication technologies for automizing the value stream mapping process with focus on the manufacturing process.

Keywords: lean management 4.0, value stream management (VSM) 4.0, dynamic value stream mapping, enterprise resource planning (ERP)

Procedia PDF Downloads 147

Authors: Rui Yin, Haojie Wang, Xun Li

Abstract:

Recently, mainstream dense retrieval methods have obtained state-of-the-art results on some datasets and tasks. However, they require large amounts of training data, which is not available in most domains. The severe performance degradation of dense retrievers on new data domains has limited the use of dense retrieval methods to only a few domains with large training datasets. In this paper, we propose an unsupervised domain-adaptive approach based on query generation. First, a generative model is used to generate relevant queries for each passage in the target corpus, and then the generated queries are used for mining negative passages. Finally, the query-passage pairs are labeled with a cross-encoder and used to train a domain-adapted dense retriever. Experiments show that our approach is more robust than previous methods in target domains that require less unlabeled data.

Keywords: dense retrieval, query generation, unsupervised training, text retrieval

Procedia PDF Downloads 71

Authors: Muhardiyan Erawan, Zaenal Mutaqin

Abstract:

Balingara River is one of the rivers with the type Gravel-Bed in Indonesia. Gravel-Bed Rivers easily deformed in a relatively short time due to several variables, that are climate (rainfall), river discharge, topography, rock types, and land cover. To determine stream channel changes in Balingara River used Landsat 7 and 8 and analyzed planimetric or two dimensions. Parameters to determine changes in the stream channel are sinuosity ratio, Brice Index, the extent of erosion and deposition. Changes in stream channel associated with changes in land cover then analyze with a descriptive analysis of spatial and temporal. The location of a stream channel has a low gradient in the upstream, and middle watershed with the type of rock in the form of gravel is more easily changed than other locations. Changes in the area of erosion and deposition influence the land cover changes.

Keywords: Brice Index, erosion, deposition, gravel-bed, land cover change, sinuosity ratio, stream channel change

Procedia PDF Downloads 326

Authors: Hamdi Sana, Emna Bouazizi, Sami Faiz

Abstract:

In recent years, there is a huge increase in the use of spatio-temporal applications where data and queries are continuously moving. As a result, the need to process real-time spatio-temporal data seems clear and real-time stream data management becomes a hot topic. Sliding window model and frequent itemset mining over dynamic data are the most important problems in the context of data mining. Thus, sliding window model for frequent itemset mining is a widely used model for data stream mining due to its emphasis on recent data and its bounded memory requirement. These methods use the traditional transaction-based sliding window model where the window size is based on a fixed number of transactions. Actually, this model supposes that all transactions have a constant rate which is not suited for real-time applications. And the use of this model in such applications endangers their performance. Based on these observations, this paper relaxes the notion of window size and proposes the use of a timestamp-based sliding window model. In our proposed frequent itemset mining algorithm, support conditions are used to differentiate frequents and infrequent patterns. Thereafter, a tree is developed to incrementally maintain the essential information. We evaluate our contribution. The preliminary results are quite promising.

Keywords: real-time spatial big data, frequent itemset, transaction-based sliding window model, timestamp-based sliding window model, weighted frequent patterns, tree, stream query

Procedia PDF Downloads 157

Authors: Simon Keller, Rainer Mueller

Abstract:

Continuous range queries are a common means to handle mobile clients in high-density areas. Most existing approaches focus on settings in which the range queries for location-based services are more or less static, whereas the mobile clients in the ranges move. We focus on a category called dynamic real-time range queries (DRRQ), assuming that both, clients requested by the query and the inquirers, are mobile. In consequence, the query parameters and the query results continuously change. This leads to two requirements: the ability to deal with an arbitrarily high number of mobile nodes (scalability) and the real-time delivery of range query results. In this paper, we present the highly decentralized solution adaptive quad streaming (AQS) for the requirements of DRRQs. AQS approximates the query results in favor of a controlled real-time delivery and guaranteed scalability. While prior works commonly optimize data structures on the involved servers, we use AQS to focus on a highly distributed cell structure without data structures automatically adapting to changing client distributions. Instead of the commonly used request-response approach, we apply a lightweight streaming method in which no bidirectional communication and no storage or maintenance of queries are required at all.

Keywords: approximation of client distributions, continuous spatial range queries, mobile objects, streaming-based decentralization in spatial mobile environments

Procedia PDF Downloads 142

Authors: Hasan Kalyoncu, Füsun Kılçık, Hatice Gülboy Akyıldırım, Aynur Özen, Mehmet Acar, Nur Yoluk

Abstract:

For evaluation of water quality of the river systems in Antalya Basin, macrozoobenthos samples were taken from 22 determined stations by a hand net and identified at family level. Water quality of Antalya Basin was determined according to Biological Monitoring Working Party (BMWP) system, by using macrozoobenthic invertebrates and physicochemical parameters. As a result of the evaluation, while Aksu Stream was determined as the most polluted stream in Antalya Basin, Isparta Stream was determined as the most polluted tributary of Aksu Stream. Pollution level of the Isparta Stream was determined as quality class V and it is the extremely polluted part of stream. Pollution loads at the sources of the streams were determined in low levels in general. Due to some parts of the streams have passed through deep canyons and take their sources from nonresidential and non-arable regions, majority of the streams that take place in Antalya Basin are at high quality level. Waste water, which comes from agricultural and residential regions, affects the lower basins of the streams. Because of the waste water, lower parts of the stream basins exposed to the pollution under anthropogenic effects. However, in Aksu Stream, which differs by being exposed to domestic and industrial wastes of Isparta City, extreme pollution was determined, particularly in the Isparta Stream part.

Keywords: Antalya basin, biomonitoring, BMWP, water quality

Procedia PDF Downloads 320

Authors: Hasan Kalyoncu, Ayşe Güneş

Abstract:

This study was carried on 12 determined stations, on Alara and Karpuz Streams, between January and November 2014. Seasonal samples were taken from the stations to analyze physicochemical parameters and Plecoptera Fauna in the water. The correlation between identified taxa and physicochemical data were tried to determine. As the result of the study, 2088 individuals from Plecoptera fauna were examined, 3 genera and 13 species were identified. The taxa of Brachyptera risi, Capnia bifrons, Dinocras cephalotes, Diura bicaudata, Isogenus nebecula, Isogenus sp., Isoperla grammatica, Leuctra hippopus, Leuctra inermis, Leuctra moselyi, Leuctra sp., Nemoura sp., Perla bipunctata, Perla marginata, Protonemura meyeri and Rhabdiopteryx acuminata were determined. In Alara Stream, the dominant species were; Isogenus nebecula at stations I and IV, Leuctra moselyi at station II, Leuctra hippopus at stations III, V and VI. In Karpuz Stream, Brachyptera risi was the dominant species in all stations. While Leuctra hippopus was the dominant taxon in Alara Stream, in Karpuz Stream it was Brachyptera risi. The highest diversity value was at station III and the lowest was at station VI in Alara Stream and the lowest diversity value was at station VI, while the highest was at station I in Karpuz Stream. In Alara Stream, the most similar stations were I and III, while in Karpuz Stream the highest similarity was determined between stations I and II. As for the evaluation result, the water quality of Alara and Karpuz Streams were determined as at oligosaprobic level.

Keywords: Alara stream, Karpuz stream, plecoptera, water quality

Procedia PDF Downloads 294

Authors: K. K. Soundra Pandian, Bhupendra Gupta

Abstract:

In the modern era, as the application data’s are massive and complex, it needs to be secured from the adversary attack. In this context, a non-recursive key based integrated spritz stream cipher with the circulant hash function using delayed feedback with carry shift register (d-FCSR) is proposed in this paper. The novelty of this proposed stream cipher algorithm is to engender the improved keystream using d-FCSR. The proposed algorithm is coded using Verilog HDL to produce dynamic binary key stream and implemented on commercially available FPGA device Virtex 5 xc5vlx110t-2ff1136. The implementation of stream cipher using d-FCSR on the FPGA device operates at a maximum frequency of 60.62 MHz. It achieved the data throughput of 492 Mbps and improved in terms of efficiency (throughput/area) compared to existing techniques. This paper also briefs the cryptanalysis of proposed circulant hash value based spritz stream cipher using d-FCSR is against the adversary attack on a hardware platform for the hardware based cryptography applications.

Keywords: cryptography, circulant function, field programmable gated array, hash value, spritz stream cipher

Procedia PDF Downloads 247

Authors: Benjamin Bobbia, Matthias Picard

Abstract:

In recent years active learning methodologies based on the representativity of the data seems more promising to limit overfitting. The presented query methodology for regression using the Wasserstein distance measuring the representativity of our labelled dataset compared to the global distribution. In this work a crucial use of GroupSort Neural Networks is made therewith to draw a double advantage. The Wasserstein distance can be exactly expressed in terms of such neural networks. Moreover, one can provide explicit bounds for their size and depth together with rates of convergence. However, heterogeneity of the dataset is also considered by weighting the Wasserstein distance with the error of approximation at the previous step of active learning. Such an approach leads to a reduction of overfitting and high prediction performance after few steps of query. After having detailed the methodology and algorithm, an empirical study is presented in order to investigate the range of our hyperparameters. The performances of this method are compared, in terms of numbers of query needed, with other classical and recent query methods on several UCI datasets.

Keywords: active learning, Lipschitz regularization, neural networks, optimal transport, regression

Procedia PDF Downloads 79

Authors: Ahlem Bouziri, Chiraz Latiri, Eric Gaussier

Abstract:

The steady growth in the size of textual document collections is a key progress-driver for modern information retrieval techniques whose effectiveness and efficiency are constantly challenged. Given a user query, the number of retrieved documents can be overwhelmingly large, hampering their efficient exploitation by the user. In addition, retaining only relevant documents in a query answer is of paramount importance for an effective meeting of the user needs. In this situation, the query expansion technique offers an interesting solution for obtaining a complete answer while preserving the quality of retained documents. This mainly relies on an accurate choice of the added terms to an initial query. Interestingly enough, query expansion takes advantage of large text volumes by extracting statistical information about index terms co-occurrences and using it to make user queries better fit the real information needs. In this respect, a promising track consists in the application of data mining methods to extract dependencies between terms, namely a generic basis of association rules between terms. The key feature of our approach is a better trade off between the size of the mining result and the conveyed knowledge. Thus, face to the huge number of derived association rules and in order to select the optimal combination of query terms from the generic basis, we propose to model the problem as a classification problem and solve it using a supervised learning algorithm such as SVM or k-means. For this purpose, we first generate a training set using a genetic algorithm based approach that explores the association rules space in order to find an optimal set of expansion terms, improving the MAP of the search results. The experiments were performed on SDA 95 collection, a data collection for information retrieval. It was found that the results were better in both terms of MAP and NDCG. The main observation is that the hybridization of text mining techniques and query expansion in an intelligent way allows us to incorporate the good features of all of them. As this is a preliminary attempt in this direction, there is a large scope for enhancing the proposed method.

Keywords: supervised leaning, classification, query expansion, association rules

Procedia PDF Downloads 321

Authors: B. Suraj Aravind, M. H. M. Krishna Prasad

Abstract:

There exist emerging applications of data streams that require association rule mining, such as network traffic monitoring, web click streams analysis, sensor data, data from satellites etc. Data streams typically arrive continuously in high speed with huge amount and changing data distribution. This raises new issues that need to be considered when developing association rule mining techniques for stream data. This paper proposes to introduce an improved data stream association rule mining algorithm by eliminating the limitation of resources. For this, the concept of cloud computing is used. Inclusion of this may lead to additional unknown problems which needs further research.

Keywords: data stream, association rule mining, cloud computing, frequent itemsets

Procedia PDF Downloads 498

Authors: Satyavati Shukla, Lakhan V. Rathod, Mohan V. Khire

Abstract:

Rapid urbanization changes the land use/land cover pattern of a developing region. Due to these land surface changes, stream flow of the rivers also changes. It is important to investigate the factors affecting hydrological characteristics of the river basin for better river basin management planning. This study is aimed to understand the effect of Land Use/Land Cover (LU/LC) changes on stream flow of Upper Bhima River basin which is highly stressed in terms of water resources. In this study, Upper Bhima River basin is divided into two adjacent sub-watersheds: Mula-Mutha (urbanized) sub-watershed and Bhima (non-urbanized) sub-watershed. First of all, LU/LC changes were estimated over 1980, 2002, and 2009 for both Mula-Mutha and Bhima sub-watersheds. Further, stream flow simulations were done using Soil and Water Assessment Tool (SWAT) for the streams draining both watersheds. Results revealed that stream flow was relatively higher for urbanized sub-watershed. Through Sensitivity Analysis it was observed that out of all the parameters used, base flow was the most sensitive parameter towards LU/LC changes.

Keywords: land use/land cover, remote sensing, stream flow, urbanization

Procedia PDF Downloads 319

Authors: Rui Yin, Haojie Wang, Xun Li

Abstract:

Recently, mainstream dense retrieval methods have obtained state-of-the-art results on some datasets and tasks. However, they require large amounts of training data, which is not available in most domains. The severe performance degradation of dense retrievers on new data domains has limited the use of dense retrieval methods to only a few domains with large training datasets. In this paper, we propose an unsupervised domain-adaptive approach based on query generation. First, a generative model is used to generate relevant queries for each passage in the target corpus, and then, the generated queries are used for mining negative passages. Finally, the query-passage pairs are labeled with a cross-encoder and used to train a domain-adapted dense retriever. We also explore contrastive learning as a method for training domain-adapted dense retrievers and show that it leads to strong performance in various retrieval settings. Experiments show that our approach is more robust than previous methods in target domains that require less unlabeled data.

Keywords: dense retrieval, query generation, contrastive learning, unsupervised training

Procedia PDF Downloads 99

Authors: Shaode Yu, Jiajian Meng, Bing Zhu, Hang Yu, Qiurui Sun

Abstract:

Speech emotion recognition (SER) is for recognizing human subjective emotions through audio data in-depth analysis. From speech audios, how to comprehensively extract emotional information and how to effectively fuse extracted features remain challenging. This paper presents a dual-stream SER framework that embraces both full training and transfer learning of different networks for thorough feature encoding. Besides, a plug-and-play cross-attention fusion (CAF) module is implemented for the valid integration of the dual-stream encoder output. The effectiveness of the proposed CAF module is compared to the other three fusion modules (feature summation, feature concatenation, and feature-wise linear modulation) on two databases (RAVDESS and IEMO-CAP) using different dual-stream encoders (full training network, DPCNN or TextRCNN; transfer learning network, HuBERT or Wav2Vec2). Experimental results suggest that the CAF module can effectively reconcile conflicts between features from different encoders and outperform the other three feature fusion modules on the SER task. In the future, the plug-and-play CAF module can be extended for multi-branch feature fusion, and the dual-stream SER framework can be widened for multi-stream data representation to improve the recognition performance and generalization capacity.

Keywords: speech emotion recognition, cross-attention fusion, dual-stream, pre-trained

Procedia PDF Downloads 72