Context Extraction

The rapid growth of Global Positioning System (GPS) and availability of real-time Geo-located data allow the mobile devices to provide information which leads towards the Location Based Services (LBS). The need for providing suggestions to personals about the activities of their interests, the LBS contributing more effectively to this purpose. Recommendation system (RS) is one of the most effective and efficient features that has been initiated by the LBS. Our proposed system is intended to design a recommendation system that will provide suggestions to the user and also find a suitable place for a group of users and it is according to their preferred type of places. In our work, we propose the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm for clustering the check-in spots of the user’s and user-based Collaborative Filtering (CF) to find similar users as we are considering constructing an interest profile for each user. We also introduced a grid-base...

Wireless Sensors enable fine grain monitoring of activities of individual and social interest. Typically these sensors sense & send data continuously directly or through other sensor nodes to a base station. Wireless Sensor Data are inherently noisy and have frequent random spikes due to dynamic nature of the medium. Hence, the decision at the receiving node based on such data is likely to be erroneous. Erroneous data and decisions may affect its transformation to meaningful form like 'context'. It is therefore desirable to clean the data for improved context extraction. Bayesian Belief Networks are used here to quantitatively encode the dependencies among various sensors. These dependencies are then used to estimate missing data and also to detect and recover from errors. Cleaned data is then used for deriving Contextual Information and it results in improved context feature calculation. In this paper five algorithms for Bayesian Belief Network Construction have been evaluated and their performance of classification studied. Conjunctive rules are defined to map the sensors to already defined context. A secondary data obtained from weather sensor boards installed at Intel research lab at Berkeley have been used to demonstrate the approach.

Google Map is where all spots right now dropped in. By deciding the present area, they can rapidly explore to the right area as per the client's place. On the off chance that somebody can't be accessible in the disconnected world contrasted with the web, and the reaction is additionally little disconnected. Here this application concentrated primarily on next-place visits dependent on the versatility logs, the interests and inclinations of clients. In this paper we proposed a new algorithm called NNTGCFEA (Neural Network Trained Global Collaborative Feature Extraction Algorithm). Our proposed Algorithm considering the Latitude and longitude, the present area can be effectively anticipated dependent on specific strategies when clients visiting previous authentic registration to the particular area dependent on frequencies and rankings. In this manner, in the event that they are close by and both are recorded on the site, it assists with seeing the companions post and remarks to a specific spot.

This paper proposes a recommender system that exploits linked open data (LOD) to perform a social context-aware cross-domain recommendation of personalized itineraries integrated with multimedia and textual content. To this aim, the recommendation engine considers the user profile, the context of use, and the features of the points of interest (POIs) extracted from LOD sources. We describe how to extract data and process it to perform hybrid filtering. All recommendations are based on the user's features extracted implicitly and explicitly. These features are used to apply content-based filtering and collaborative filtering, weighing results based on similar users' experience. The preliminary results of an experimental evaluation on a sample of 20 real users show the effectiveness of the proposed system not only in terms of perceived accuracy, but also in terms of novelty, non-obviousness, and serendipity.

CNN Universal Machines that contain two different processors working interactively with each other, have an important impact for image processing applications with their advanced computing features. These processors are named as the ACE16k processor which is the hardware implementation of cellular neural networks and the Digital Signal Processor (DSP). Bio-inspired (Bi-i) Cellular Vision System is also a CNN Universal Machine and its standalone architecture is built on CNN-type (ACE16k) and DSP-type microprocessors. In this study, certain objects in moving images are detected and their features are extracted. By using these features, an algorithm that finds out the path of moving vehicles and determines the traffic density is implemented on the Bi-i Cellular Vision System. Finally, the output images and the data related with traffic density which are obtained as results of this implementation are evaluated.

Predicting the location of a mobile user in the near future can be used for a large number of user-centered ubiquitous applications. This can be extended to crowdcentered applications if a large number of users is included. In this paper we present a spatio-temporal prediction approach to forecast user location in a medium-term period. Our approach is based on the hypothesis that users exhibit a different mobility pattern for each day of the week. Once factored out this weekly pattern, user mobility among points of interest is postulated to be markovian. We trained a hidden Markov model to forecast user mobility and evaluated our approach using a public dataset. The experimental results show that our approach is effective considering a time period of up to seven hours. We obtained an accuracy of up to 81.75 % for a period of 30 minutes, and 66.25 % considering 7 hours.

The usual approach to named-entity detection is to learn extraction rules that rely on linguistic, syntactic, or document format patterns that are consistent across a set of documents. However, when there is no consistency among documents, it may be more effective to learn document-specific extraction rules. This paper presents a knowledge-based approach to learning rules for named-entity extraction. Documentspecific extraction rules are created using a generate-andtest paradigm and a database of known named-entities. Experimental results show that this approach is effective on Web documents that are difficult for the usual methods.

Google Map is where all spots right now dropped in. By deciding the present area, they can rapidly explore to the right area as per the client's place. On the off chance that somebody can't be accessible in the disconnected world contrasted with the web, and the reaction is additionally little disconnected. Here this application concentrated primarily on next-place visits dependent on the versatility logs, the interests and inclinations of clients. In this paper we proposed a new algorithm called NNTGCFEA (Neural Network Trained Global Collaborative Feature Extraction Algorithm). Our proposed Algorithm considering the Latitude and longitude, the present area can be effectively anticipated dependent on specific strategies when clients visiting previous authentic registration to the particular area dependent on frequencies and rankings. In this manner, in the event that they are close by and both are recorded on the site, it assists with seeing the companions post and remarks to a specific spot.

Abstract—Despite of the hope arised a few years ago, Content Based Image Retrieval-CBIR-systems has not reached the initial goal, ie to manage and search images in database: we are unable to link the semantic sens of an image to numerical values. However, some ...

CNN Universal Machines that contain two different processors working interactively with each other, have an important impact for image processing applications with their advanced computing features. These processors are named as the ACE16k processor which is the hardware implementation of cellular neural networks and the Digital Signal Processor (DSP). Bio-inspired (Bi-i) Cellular Vision System is also a CNN Universal Machine and its standalone architecture is built on CNN-type (ACE16k) and DSP-type microprocessors. In this study, certain objects in moving images are detected and their features are extracted. By using these features, an algorithm that finds out the path of moving vehicles and determines the traffic density is implemented on the Bi-i Cellular Vision System. Finally, the output images and the data related with traffic density which are obtained as results of this implementation are evaluated.

Abstract—At this time all people, especially managers and businessmen, are exposed to the ever-present information pollution. This is why tools of business intelligence are of great importance; nevertheless the current methods can hardly cope with large and unstructured text sources like World Wide Web that currently becomes more and more important. To achieve this main goal we have to find and verify satisfactorily reliable methods for automatic extraction of a main context of a document, ie, multidimensional structured ...

In this paper, the development of informatics tools and their testing for active learning support is described. The tools -programming scripts, optional menus, and tailored applications, are a part of the developed "Batch Information and Knowledge Editor". They are selected by a teacher from the knowledge editor's menu, and in the classroom are used directly for performing active learning at the university's detached workplace.

This paper introduces Named Entity Recognition approach for text corpus. Supervised Statistical methods are used to develop our system. Our system can be used to categorize NEs belonging to a particular domain for which it is being trained. As Named Entities appears in text surrounded by contexts (words that are left or right of the NE), we will be focusing on extracting NE contexts from text and then performing statistical computing on them. We are using n-gram model for extracting contexts from text. Our methodology first extracts left and right tri-grams surrounding NE instances in the training corpus and calculate their probabilities. Then all the extracted trigrams along with their calculated probabilities are stored in a file. During testing, system detects unrecognized NEs from the testing corpus and categorizes them using the tri-gram probabilities calculated during training time. The proposed system is made up of two modules i.e. Knowledge acquisition and NE Recognition. Kn...

This paper introduces Named Entity Recognition approach for text corpus. Supervised Statistical methods are used to develop our system. Our system can be used to categorize NEs belonging to a particular domain for which it is being trained. As Named Entities appears in text surrounded by contexts (words that are left or right of the NE), we will be focusing on extracting NE contexts from text and then performing statistical computing on them. We are using n-gram model for extracting contexts from text. Our methodology first extracts left and right tri-grams surrounding NE instances in the training corpus and calculate their probabilities. Then all the extracted trigrams along with their calculated probabilities are stored in a file. During testing, system detects unrecognized NEs from the testing corpus and categorizes them using the tri-gram probabilities calculated during training time. The proposed system is made up of two modules i.e. Knowledge acquisition and NE Recognition. Kn...

An increasing safety and reducing road accidents, thereby by saving lives are one of great interest in the context of Advanced Driver Assistance Systems. Apparently, among the complex and challenging tasks of future road vehicles is road lane detection or even road boundaries detection. We present a robust and real time approach to lane marker detection in urban streets. A lane-detection method aimed at handling moving vehicles in the traffic scenes is proposed in this brief. First, an adaptive region of interest ROI is set. The ROI is mainly in the bottom half of the image since the main lane information only appears in the bottom of the image. Then the lane marks are extracted based on color information. The extraction of lane-mark colors is designed in a way that is not affected by illumination changes and the proportion of space that vehicles on the road occupy.

The problem of language barrier in medical consultations is a limitation to healthcare provisioning to the larger population of the developing and underdeveloped world. The doctor to patient ratio is dismally low, especially in cases where patients can only converse in their native languages. Hence there is the need to provide solutions to the problem of language barrier in medical consultations. This work considers the over 250 ethnic communities with more than 350 local languages and dialects in Nigeria as a serious challenge knowing that a great percentage of this population are semi to stack illiterates. A multilingual application is developed to support information dissemination to multiple local languages in Nigeria while also supporting communication between medical practitioners and indigenous patient with natural language barrier. This enables an online real time medical consultation on mobile devices. A working model is designed for a language translator (English to Nigeri...

The problem of language barrier in medical consultations is a limitation to healthcare provisioning to the larger population of the developing and underdeveloped world. The doctor to patient ratio is dismally low, especially in cases where patients can only converse in their native languages. Hence there is the need to provide solutions to the problem of language barrier in medical consultations. This work considers the over 250 ethnic communities with more than 350 local languages and dialects in Nigeria as a serious challenge knowing that a great percentage of this population are semi to stack illiterates. A multilingual application is developed to support information dissemination to multiple local languages in Nigeria while also supporting communication between medical practitioners and indigenous patient with natural language barrier. This enables an online real time medical consultation on mobile devices. A working model is designed for a language translator (English to Nigeria major languages: Hausa, Igbo and Yoruba) for a mobile medical consultation making use of various components such as the mobile client terminal, the internet application server etc. The resulting solution was subjected to evaluation by medical experts to ascertain the relevance of the solution in the current state of clinical solutions research. Results from this work show that patients who are not educated in the official language can communicate with clinicians of different lingua background.

Despite of the hope arised a few years ago, Content Based Image Retrieval CBIR systems has not reached the initial goal, ie to manage and search images in database: we are unable to link the semantic sens of an image to numerical values. However, some members of the community have begun the necessary introspection. The analyze of each step of the feature extraction will allow us to overcome actual problematics and to take the right path in the future. In this context, we propose in this paper to discuss about a low-level tool frequently used: the segmentation step. In the general context of scene images, we evaluate the stability of some classical algorithms using a basic protocol. The quite inefficiency of all approaches let us conclude to the necessity to use meta-data and any other collected informations during this first segmentation step.

In this paper, the development of informatics tools and their testing for active learning support is described. The tools-programming scripts, optional menus, and tailored applications, are a part of the developed "Batch Information and Knowledge Editor". They are selected by a teacher from the knowledge editor's menu, and in the classroom are used directly for performing active learning at the university's detached workplace.

We have been witnessing a meticulous expansion in the amount of biological databases as an outcome of the human genome-sequencing project. These biological databases are created and updated by the inventions of new molecules by the biologists. The nature of most of these databases is either non-structured or semi-structured. The data are stored in a flat file, which makes it difficult to retrieve a particular record in reasonable time. Computational biology tasks such as multiple sequence alignment, sequence similarity, motif finding, and structure prediction have yanked many researchers. We feel that computational biologists are many a time not interested in all the fields present in the database. Rather, they are concerned about particular fields depending upon the issue being addressed. We have developed utilities to extract and index UniRef100 database for fast sequential and indexed random access, to normalize occurrences of pairs, trios and quads substrings of amino acids in the database, a programmatically mutated database to test the sequence similarity algorithms. This work shall aid the upcoming researchers in the field of computational biology to customize existing database for the algorithmic needs to accelerate the operations.

We have been witnessing a meticulous expansion in the amount of biological databases as an outcome of the human genome-sequencing project. These biological databases are created and updated by the inventions of new molecules by the biologists. The nature of most of these databases is either non-structured or semi-structured. The data are stored in a flat file, which makes it difficult to retrieve a particular record in reasonable time. Computational biology tasks such as multiple sequence alignment, sequence similarity, motif finding, and structure prediction have yanked many researchers. We feel that computational biologists are many a time not interested in all the fields present in the database. Rather, they are concerned about particular fields depending upon the issue being addressed. We have developed utilities to extract and index UniRef100 database for fast sequential and indexed random access, to normalize occurrences of pairs, trios and quads substrings of amino acids in the database, a programmatically mutated database to test the sequence similarity algorithms. This work shall aid the upcoming researchers in the field of computational biology to customize existing database for the algorithmic needs to accelerate the operations.

In this paper, the development of informatics tools and their testing for active learning support is described. The tools -programming scripts, optional menus, and tailored applications, are a part of the developed "Batch Information and Knowledge Editor". They are selected by a teacher from the knowledge editor's menu, and in the classroom are used directly for performing active learning at the university's detached workplace.

Customer Relationship Management has become the main interest of researchers and practitioners especially in the domains of Marketing and Information Systems (IS). This paper is an overview on success factors that could facilitate successful implementation of CRM. Index Terms-information systems, information technology, CRM components, CRM success factors, customer relationship management.

Abstract—Despite of the hope arised a few years ago, Content Based Image Retrieval-CBIR-systems has not reached the initial goal, ie to manage and search images in database: we are unable to link the semantic sens of an image to numerical values. However, some ...

Abstract—Despite of the hope arised a few years ago, Content Based Image Retrieval-CBIR-systems has not reached the initial goal, ie to manage and search images in database: we are unable to link the semantic sens of an image to numerical values. However, some ...

The problem of language barrier in medical consultations is a limitation to healthcare provisioning to the larger population of the developing and underdeveloped world. The doctor to patient ratio is dismally low, especially in cases where patients can only converse in their native languages. Hence there is the need to provide solutions to the problem of language barrier in medical consultations. This work considers the over 250 ethnic communities with more than 350 local languages and dialects in Nigeria as a serious challenge knowing that a great percentage of this population are semi to stack illiterates.

Recent research on geographical data mining that focuses on user behavior is lacking some fundamental aspects; measurements rely on large quantities of geographic data and lack contextual information. This work introduces a novel knowledge discovery architecture that brings together machine learning techniques with readily available information from popular Location Social Networks, in order to enrich geographical locations with context and add meaning to user behavior. Results show that through analysis of context enriched data we are capable of inferring context for detected user points of interest and patterns, such as where the user lives, works and spends his free time, without a large quantity of information or prior knowledge of the user and his private data.

The problem of language barrier in medical consultations is a limitation to healthcare provisioning to the larger population of the developing and underdeveloped world. The doctor to patient ratio is dismally low, especially in cases where patients can only converse in their native languages. Hence there is the need to provide solutions to the problem of language barrier in medical consultations. This work considers the over 250 ethnic communities with more than 350 local languages and dialects in Nigeria as a serious challenge knowing that a great percentage of this population are semi to stack illiterates.

Dr. Cecilia Mascolo is a Reader in Mobile Systems in the Computer Laboratory, University of Cambridge. Her research interests are in human mobility modeling. She has published in the areas of mobile computing, mobility and social modeling and sensor networking. She is and has been PI and co-PI of EPSRC projects on mobile and sensor networks (SATIN, CREAM, FRESNEL, UBhave, UBiVal and WILDSENSING) and on EPSRC project MOLTEN, investigating time varying complex networks.

At this time all people, especially managers and businessmen, are exposed to the ever-present information pollution. This is why tools of business intelligence are of great importance; nevertheless the current methods can hardly cope with large and unstructured text sources like World Wide Web that currently becomes more and more important. To achieve this main goal we have to find and verify satisfactorily reliable methods for automatic extraction of a main context of a document, i.e., multidimensional structured characterization representing the main topic of the document. To cope with the multilingual sources we have to develop approaches that would not be dependent on the language of the source and that would not need any additional language dependent tools (like thesauri). In our conception, the context is dynamic – it means that a classification of a document will not be dependent only on the document in question but also on the corpus; the expansion of a corpus can result in a change of a document classification.