Bernabé Batchakui - Academia.edu (original) (raw)

Papers by Bernabé Batchakui

World journal of engineering and technology, 2024

The advancement in technology accelerated and opened availability of various alternatives to make... more The advancement in technology accelerated and opened availability of various alternatives to make a choice in every domain. In the era of big data it is a tedious and time consuming task to evaluate the features of a large amount of information provided to make a choice. One solution to ease this overload problem is building recommender system that can process a large amount of data and support users’ decision making ability. In this paper different traditional recommendation techniques, deep learning approaches for recommender system and survey of deep learning techniques on recommender system are presented. A variety of techniques have been proposed to perform recommendation, including content based, collaborative and hybrid recommenders. Due to the limitation of the traditional recommendation methods in obtaining accurate result a deep learning approach is introduced both for collaborative and content based

World Academy of Science, Engineering and Technology, International Journal of Mathematical and Computational Sciences, Sep 21, 2017

Pédagogies en développement, Sep 1, 2022

arXiv (Cornell University), Feb 20, 2023

In this paper, we propose an alternative to deep neural networks for semantic information retriev... more In this paper, we propose an alternative to deep neural networks for semantic information retrieval for the case of long documents. This new approach exploiting clustering techniques to take into account the meaning of words in Information Retrieval systems targeting long as well as short documents. This approach uses a specially designed clustering algorithm to group words with similar meanings into clusters. The dual representation (lexical and semantic) of documents and queries is based on the vector space model proposed by Gerard Salton in the vector space constituted by the formed clusters. The originalities of our proposal are at several levels: first, we propose an efficient algorithm for the construction of clusters of semantically close words using word embeddings as input, then we define a formula for weighting these clusters, and then we propose a function allowing to combine efficiently the meanings of words with a lexical model widely used in Information Retrieval. The evaluation of our proposal in three contexts with two different datasets (SQuAD and TREC-CAR) has shown that is significantly improves the classical approaches only based on the keywords without degrading the lexical aspect.

2020 IEEE Learning With MOOCS (LWMOOCS), 2020

MOOCs (Massive Open Online Courses) are definitely one of the best approach to support the intern... more MOOCs (Massive Open Online Courses) are definitely one of the best approach to support the international agenda about inclusive and equitable education and lifelong learning opportunities for all (SDG4) [1]. A great deal universities and institutions offer valuable free courses to their numerous students and to people around the word through MOOC platforms. However, because of huge number of learners and data generated, learner’s behaviour in those platforms remain a kind of black box for learners themselves and for courses instructors who are supposed to tutor or monitor learners in the learning process. Therefore, learner do not receive sufficient support from instructors and from their peers, during the learning process [2]. This is one the main reasons that lead to high dropout, low completion and success rate observed in the MOOCs. Many research work have suggested descriptive, predictive and prescriptive models to address this issue, but most of these models focus on predicting dropout, completion and/or success, and do not generally pay enough attention to one of the key step (learner behaviour), that comes before, and can explain dropping out and failure. Our research aims to develop a deep learning model to predict learner behaviour (learner interactions) in the learning process, in order to equip learners and course instructors with insight understanding of the learner behaviour in the learning process. This specific paper will focus on analysing relevant algorithms to develop such model. For this analysis, we used data from UNESCO-IICBA (International Institute for Capacity Building in Africa) MOOC platform, designed for teacher training in Africa, and then we examine many types of features: geographical, social behavioural and learning behavioural features. Learner’s behaviour being a time series Big data, we built the predictive model using Deep Learning algorithms for better performance and accuracy (Thanks to the power of deep learning) compared to baseline Machine learning algorithms. Time series data is best handled by recurrent neural networks (RNN) [3], so, we choose RNN and implemented/tested the three main architectures of RNN: Simple RNNs, GRU (Gated Recurrent Unit) RNNs and LSTM (Long short-term memory) RNNs. The models were trained using L2 Regularization, based on the predictions results, we unexpectedly found model with simple RNNs produced the best performance and accuracy on the dataset used than the other RNN architectures. We had couple of observations, example: we saw a correlation between video viewing and quiz behaviour and the participation of the learner to the learning process. This observation could allow teachers to provide meaningful support and guidance to at risk or less active students. We also observed that, the shorter the video or the quiz, the more the viewer. We conclude that we could use learner video or quiz viewing behaviour to predict his behaviour concerning other MOOC contents. These results suggest the need of deeper research on educational video and educational quiz designing for MOOCs.

In the context of distance education, the challenges are: access to training platforms for mobile... more In the context of distance education, the challenges are: access to training platforms for mobile content and compatibility with the profile and the background of learners. The contribution referred in this paper is to extend VeSMEL (1) for the management of structured contents. Indeed, VeSMEL is a solution based on the VeSMp (2) protocol and the GSM network. It allows users of areas that do not have Internet to access to e-Learning platforms using their mobile phone.

International Journal of Computer Applications, 2017

The advancement in technology accelerated and opened availability of various alternatives to make... more The advancement in technology accelerated and opened availability of various alternatives to make a choice in every domain. In the era of big data it is a tedious and time consuming task to evaluate the features of a large amount of information provided to make a choice. One solution to ease this overload problem is building recommender system that can process a large amount of data and support users' decision making ability. In this paper different traditional recommendation techniques, deep learning approaches for recommender system and survey of deep learning techniques on recommender system are presented. A variety of techniques have been proposed to perform recommendation, including content based, collaborative and hybrid recommenders. Due to the limitation of the traditional recommendation methods in obtaining accurate result a deep learning approach is introduced both for collaborative and content based approaches that will enable the model to learn different features of users and items automatically to improve accuracy of recommendation. Even though deep learning poses a great impact in various areas, applying the model to a recommender systems have not been fully exploited. With the help of the advantage of deep learning in modeling different types of data, deep recommender systems can better understand users' demand to further improve quality of recommendation.

Proceedings of the Sixteenth European Conference on Computer Systems, 2021

Cloud applications based on the "Functions as a Service" (FaaS) paradigm have become ve... more Cloud applications based on the "Functions as a Service" (FaaS) paradigm have become very popular. Yet, due to their stateless nature, they must frequently interact with an external data store, which limits their performance. To mitigate this issue, we introduce OFC, a transparent, vertically and horizontally elastic in-memory caching system for FaaS platforms, distributed over the worker nodes. OFC provides these benefits cost-effectively by exploiting two common sources of resource waste: (i) most cloud tenants overprovision the memory resources reserved for their functions because their footprint is non-trivially input-dependent and (ii) FaaS providers keep function sandboxes alive for several minutes to avoid cold starts. Using machine learning models adjusted for typical function input data categories (e.g., multimedia formats), OFC estimates the actual memory resources required by each function invocation and hoards the remaining capacity to feed the cache. We build ...

International Journal of Emerging Technologies in Learning (iJET), 2022

The web is one of the primary sources of information for finding learning oriented documents. In ... more The web is one of the primary sources of information for finding learning oriented documents. In addition, the main suitable way to find information and documents on the Internet is by using search engines. Search engines are constantly improving in terms of selection algorithms and in terms of the Human Machine interface (HMI). Also, these search engines are the basis of a new field of research called Search-As-Learning. The Search-As-Learning explores information search environments to enhance learning during user search tasks. This work focuses on our view of the state of the art in the field of Search Engines in learning context and Search-As-Learning, stressing on the most recent research. We conclude by highlighting the current shortcomings on improvement of the learning aspect within search engines, and present next work which will be the association of a layer above the traditional search engines to promote the appropriation of content during search task for a learning context

Malaria affects hundreds of millions of people in the world, particularly in the tropics. This re... more Malaria affects hundreds of millions of people in the world, particularly in the tropics. This results in particularly high death rates among children and pregnant women, especially the poor living in squalid conditions. This situation is not only due to the increasing drug resistance of malaria and the resistance of the main vector to pesticide control, but also the lack of awareness on the part of communities to fight the disease. In this view, vector control is a cornerstone of the strategy that needs to be implemented and monitored. In this regard, the problem of estimating malarial transmission rate is still very important. It is the aim of this paper.

In many developing countries, the rate of success in the universities as in the high-schools rema... more In many developing countries, the rate of success in the universities as in the high-schools remains very weak and especially very disproportionate. For example for the same national or international examination the rate of success is above the average for certain institutes or establishments and below for others. Moreover the uses of communication and information technologies are increasingly numerous and offer to their users greater possibilities of collaboration and improvement of their performances. Today, the emergence of technologies of grid opens new perspectives which mitigate the limits of the current web services; the users from their working station can carry out applications on a whole of distributed and heterogeneous resources. This article is based on the infrastructure of Grid Learning Services (GLS) existing to implement an environment of collaboration which will allow the communities which have common objectives in the training area to gather, and therefore to make ...

Ontologies are becoming nowadays a recurrent topic in many research areas of computer science. Am... more Ontologies are becoming nowadays a recurrent topic in many research areas of computer science. Among other things, they are used as best model for representing knowledge of a given field of expertise. More specifically, the evolution of Semantic Web mostly relies on the use of ontologies in describing various web resources. However, the world is changing, so do resources and knowledge. Consequently, ontologies on which are based knowledge and resources are also called to evolve. One of the solutions helping to ensure this evolution is to manage the development process in order to guarantee supported services as well as some quality requirements. Many tools have been designed to ensure designing/editing of ontologies, but none of them has enough functions to ensure their evolution. Evolution of ontology considered as its adaptation to changes in the field is a key issue in the ontology engineering field. Several studies have been focused on this problem, but proposed solutions did not manage all the side effects that may derive from the evolution. This article proposes an extension of MS-ONTO [1] called POOE. POOE not only supports the evolution effects on a double plan, externally and internally (verification of integrity constraints, quality analysis of the evolved ontology) but it constitutes a true validation tool of a change log.

Journal of Higher Education Theory and Practice

In several training institutions in sub-Saharan Africa today, the competencybased approach to tea... more In several training institutions in sub-Saharan Africa today, the competencybased approach to teaching (CBA) has been adopted at the secondary school level. In Cameroon, based on our experience in teaching, we have found that this approach does not suit all categories of learners, generally the youngest. With the advent of Information and Communication Technologies (ICT), learners spend most of their time on ICT's gadgets (mobile phone, tablet, etc ...). In this paper, we propose a complement to the CBA approach through pedagogic differentiation. This differentiation takes into account the learner's environment and adds a playful and captivating aspect to the techno-pedagogic tools to be made available to them through the gadgets they use. We call this approach contextual immersion. It starts from real life situation familiar to the learner. The tool made available to the learner, which integrates this approach, guides him/her progressively towards the solution to the problem posed and a generalization that summarizes the course that will be transmitted.

International Journal of Emerging Technologies in Learning (iJET)

This paper proposes a traces model in the form of an object or class model (in the UML sense) whi... more This paper proposes a traces model in the form of an object or class model (in the UML sense) which allows the automatic calculation of indicators of various kinds and independently of the computer environment for human learning (CEHL). The model is based on the establishment of a trace-based system that encompasses all the logic of traces collecting and indicators calculation. It is im-plemented in the form of a trace database. It is an important contribution in the field of the exploitation of the traces of apprenticeship in a CEHL because it pro-vides a general formalism for modeling the traces and allowing the calculation of several indicators at the same time. Also, with the inclusion of calculated indica-tors as potential learning traces, our model provides a formalism for classifying the various indicators in the form of inheritance relationships, which promotes the reuse of indicators already calculated. Economically, the model can allow organi-zations with different learnin...

Proceedings of the 13th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management

International Journal of Recent Contributions from Engineering, Science & IT (iJES), 2015

For the past five years, the bitcoin network constantly experience a growth in its size as more c... more For the past five years, the bitcoin network constantly experience a growth in its size as more communities turn to accept the currency for payment exchanges. Using Flat File and a LevelDB of indices to save blocks on disk, bitcoin users require more memory to save the history of transaction. We focus on issues of memory management and access time in the bitcoin protocol using SQLite DataBase. With all the advantages of SQLite DataBase, it would be efficient if it is fitted in this architecture. The SQLite comes with many flavors one of which is its ability to support sql queries. Thus, instead of parsing indices to search a block from the database, a more powerful query can do the job.

Proceedings of the 4th International Conference on Theory and Practice of Electronic Governance - ICEGOV '10, 2010

In this paper, we propose VeSMDiag -- a Very Short Message Protocol based system for Medical Diag... more In this paper, we propose VeSMDiag -- a Very Short Message Protocol based system for Medical Diagnosis. VeSMDiag is a tool for enabling remote access to the medical diagnosis of first level through mobile phones.

World journal of engineering and technology, 2024

The advancement in technology accelerated and opened availability of various alternatives to make... more The advancement in technology accelerated and opened availability of various alternatives to make a choice in every domain. In the era of big data it is a tedious and time consuming task to evaluate the features of a large amount of information provided to make a choice. One solution to ease this overload problem is building recommender system that can process a large amount of data and support users’ decision making ability. In this paper different traditional recommendation techniques, deep learning approaches for recommender system and survey of deep learning techniques on recommender system are presented. A variety of techniques have been proposed to perform recommendation, including content based, collaborative and hybrid recommenders. Due to the limitation of the traditional recommendation methods in obtaining accurate result a deep learning approach is introduced both for collaborative and content based

World Academy of Science, Engineering and Technology, International Journal of Mathematical and Computational Sciences, Sep 21, 2017

Pédagogies en développement, Sep 1, 2022

arXiv (Cornell University), Feb 20, 2023

In this paper, we propose an alternative to deep neural networks for semantic information retriev... more In this paper, we propose an alternative to deep neural networks for semantic information retrieval for the case of long documents. This new approach exploiting clustering techniques to take into account the meaning of words in Information Retrieval systems targeting long as well as short documents. This approach uses a specially designed clustering algorithm to group words with similar meanings into clusters. The dual representation (lexical and semantic) of documents and queries is based on the vector space model proposed by Gerard Salton in the vector space constituted by the formed clusters. The originalities of our proposal are at several levels: first, we propose an efficient algorithm for the construction of clusters of semantically close words using word embeddings as input, then we define a formula for weighting these clusters, and then we propose a function allowing to combine efficiently the meanings of words with a lexical model widely used in Information Retrieval. The evaluation of our proposal in three contexts with two different datasets (SQuAD and TREC-CAR) has shown that is significantly improves the classical approaches only based on the keywords without degrading the lexical aspect.

2020 IEEE Learning With MOOCS (LWMOOCS), 2020

MOOCs (Massive Open Online Courses) are definitely one of the best approach to support the intern... more MOOCs (Massive Open Online Courses) are definitely one of the best approach to support the international agenda about inclusive and equitable education and lifelong learning opportunities for all (SDG4) [1]. A great deal universities and institutions offer valuable free courses to their numerous students and to people around the word through MOOC platforms. However, because of huge number of learners and data generated, learner’s behaviour in those platforms remain a kind of black box for learners themselves and for courses instructors who are supposed to tutor or monitor learners in the learning process. Therefore, learner do not receive sufficient support from instructors and from their peers, during the learning process [2]. This is one the main reasons that lead to high dropout, low completion and success rate observed in the MOOCs. Many research work have suggested descriptive, predictive and prescriptive models to address this issue, but most of these models focus on predicting dropout, completion and/or success, and do not generally pay enough attention to one of the key step (learner behaviour), that comes before, and can explain dropping out and failure. Our research aims to develop a deep learning model to predict learner behaviour (learner interactions) in the learning process, in order to equip learners and course instructors with insight understanding of the learner behaviour in the learning process. This specific paper will focus on analysing relevant algorithms to develop such model. For this analysis, we used data from UNESCO-IICBA (International Institute for Capacity Building in Africa) MOOC platform, designed for teacher training in Africa, and then we examine many types of features: geographical, social behavioural and learning behavioural features. Learner’s behaviour being a time series Big data, we built the predictive model using Deep Learning algorithms for better performance and accuracy (Thanks to the power of deep learning) compared to baseline Machine learning algorithms. Time series data is best handled by recurrent neural networks (RNN) [3], so, we choose RNN and implemented/tested the three main architectures of RNN: Simple RNNs, GRU (Gated Recurrent Unit) RNNs and LSTM (Long short-term memory) RNNs. The models were trained using L2 Regularization, based on the predictions results, we unexpectedly found model with simple RNNs produced the best performance and accuracy on the dataset used than the other RNN architectures. We had couple of observations, example: we saw a correlation between video viewing and quiz behaviour and the participation of the learner to the learning process. This observation could allow teachers to provide meaningful support and guidance to at risk or less active students. We also observed that, the shorter the video or the quiz, the more the viewer. We conclude that we could use learner video or quiz viewing behaviour to predict his behaviour concerning other MOOC contents. These results suggest the need of deeper research on educational video and educational quiz designing for MOOCs.

In the context of distance education, the challenges are: access to training platforms for mobile... more In the context of distance education, the challenges are: access to training platforms for mobile content and compatibility with the profile and the background of learners. The contribution referred in this paper is to extend VeSMEL (1) for the management of structured contents. Indeed, VeSMEL is a solution based on the VeSMp (2) protocol and the GSM network. It allows users of areas that do not have Internet to access to e-Learning platforms using their mobile phone.

International Journal of Computer Applications, 2017

The advancement in technology accelerated and opened availability of various alternatives to make... more The advancement in technology accelerated and opened availability of various alternatives to make a choice in every domain. In the era of big data it is a tedious and time consuming task to evaluate the features of a large amount of information provided to make a choice. One solution to ease this overload problem is building recommender system that can process a large amount of data and support users' decision making ability. In this paper different traditional recommendation techniques, deep learning approaches for recommender system and survey of deep learning techniques on recommender system are presented. A variety of techniques have been proposed to perform recommendation, including content based, collaborative and hybrid recommenders. Due to the limitation of the traditional recommendation methods in obtaining accurate result a deep learning approach is introduced both for collaborative and content based approaches that will enable the model to learn different features of users and items automatically to improve accuracy of recommendation. Even though deep learning poses a great impact in various areas, applying the model to a recommender systems have not been fully exploited. With the help of the advantage of deep learning in modeling different types of data, deep recommender systems can better understand users' demand to further improve quality of recommendation.

Proceedings of the Sixteenth European Conference on Computer Systems, 2021

Cloud applications based on the "Functions as a Service" (FaaS) paradigm have become ve... more Cloud applications based on the "Functions as a Service" (FaaS) paradigm have become very popular. Yet, due to their stateless nature, they must frequently interact with an external data store, which limits their performance. To mitigate this issue, we introduce OFC, a transparent, vertically and horizontally elastic in-memory caching system for FaaS platforms, distributed over the worker nodes. OFC provides these benefits cost-effectively by exploiting two common sources of resource waste: (i) most cloud tenants overprovision the memory resources reserved for their functions because their footprint is non-trivially input-dependent and (ii) FaaS providers keep function sandboxes alive for several minutes to avoid cold starts. Using machine learning models adjusted for typical function input data categories (e.g., multimedia formats), OFC estimates the actual memory resources required by each function invocation and hoards the remaining capacity to feed the cache. We build ...

International Journal of Emerging Technologies in Learning (iJET), 2022

The web is one of the primary sources of information for finding learning oriented documents. In ... more The web is one of the primary sources of information for finding learning oriented documents. In addition, the main suitable way to find information and documents on the Internet is by using search engines. Search engines are constantly improving in terms of selection algorithms and in terms of the Human Machine interface (HMI). Also, these search engines are the basis of a new field of research called Search-As-Learning. The Search-As-Learning explores information search environments to enhance learning during user search tasks. This work focuses on our view of the state of the art in the field of Search Engines in learning context and Search-As-Learning, stressing on the most recent research. We conclude by highlighting the current shortcomings on improvement of the learning aspect within search engines, and present next work which will be the association of a layer above the traditional search engines to promote the appropriation of content during search task for a learning context

Malaria affects hundreds of millions of people in the world, particularly in the tropics. This re... more Malaria affects hundreds of millions of people in the world, particularly in the tropics. This results in particularly high death rates among children and pregnant women, especially the poor living in squalid conditions. This situation is not only due to the increasing drug resistance of malaria and the resistance of the main vector to pesticide control, but also the lack of awareness on the part of communities to fight the disease. In this view, vector control is a cornerstone of the strategy that needs to be implemented and monitored. In this regard, the problem of estimating malarial transmission rate is still very important. It is the aim of this paper.

In many developing countries, the rate of success in the universities as in the high-schools rema... more In many developing countries, the rate of success in the universities as in the high-schools remains very weak and especially very disproportionate. For example for the same national or international examination the rate of success is above the average for certain institutes or establishments and below for others. Moreover the uses of communication and information technologies are increasingly numerous and offer to their users greater possibilities of collaboration and improvement of their performances. Today, the emergence of technologies of grid opens new perspectives which mitigate the limits of the current web services; the users from their working station can carry out applications on a whole of distributed and heterogeneous resources. This article is based on the infrastructure of Grid Learning Services (GLS) existing to implement an environment of collaboration which will allow the communities which have common objectives in the training area to gather, and therefore to make ...

Ontologies are becoming nowadays a recurrent topic in many research areas of computer science. Am... more Ontologies are becoming nowadays a recurrent topic in many research areas of computer science. Among other things, they are used as best model for representing knowledge of a given field of expertise. More specifically, the evolution of Semantic Web mostly relies on the use of ontologies in describing various web resources. However, the world is changing, so do resources and knowledge. Consequently, ontologies on which are based knowledge and resources are also called to evolve. One of the solutions helping to ensure this evolution is to manage the development process in order to guarantee supported services as well as some quality requirements. Many tools have been designed to ensure designing/editing of ontologies, but none of them has enough functions to ensure their evolution. Evolution of ontology considered as its adaptation to changes in the field is a key issue in the ontology engineering field. Several studies have been focused on this problem, but proposed solutions did not manage all the side effects that may derive from the evolution. This article proposes an extension of MS-ONTO [1] called POOE. POOE not only supports the evolution effects on a double plan, externally and internally (verification of integrity constraints, quality analysis of the evolved ontology) but it constitutes a true validation tool of a change log.

Journal of Higher Education Theory and Practice

In several training institutions in sub-Saharan Africa today, the competencybased approach to tea... more In several training institutions in sub-Saharan Africa today, the competencybased approach to teaching (CBA) has been adopted at the secondary school level. In Cameroon, based on our experience in teaching, we have found that this approach does not suit all categories of learners, generally the youngest. With the advent of Information and Communication Technologies (ICT), learners spend most of their time on ICT's gadgets (mobile phone, tablet, etc ...). In this paper, we propose a complement to the CBA approach through pedagogic differentiation. This differentiation takes into account the learner's environment and adds a playful and captivating aspect to the techno-pedagogic tools to be made available to them through the gadgets they use. We call this approach contextual immersion. It starts from real life situation familiar to the learner. The tool made available to the learner, which integrates this approach, guides him/her progressively towards the solution to the problem posed and a generalization that summarizes the course that will be transmitted.

International Journal of Emerging Technologies in Learning (iJET)

This paper proposes a traces model in the form of an object or class model (in the UML sense) whi... more This paper proposes a traces model in the form of an object or class model (in the UML sense) which allows the automatic calculation of indicators of various kinds and independently of the computer environment for human learning (CEHL). The model is based on the establishment of a trace-based system that encompasses all the logic of traces collecting and indicators calculation. It is im-plemented in the form of a trace database. It is an important contribution in the field of the exploitation of the traces of apprenticeship in a CEHL because it pro-vides a general formalism for modeling the traces and allowing the calculation of several indicators at the same time. Also, with the inclusion of calculated indica-tors as potential learning traces, our model provides a formalism for classifying the various indicators in the form of inheritance relationships, which promotes the reuse of indicators already calculated. Economically, the model can allow organi-zations with different learnin...

Proceedings of the 13th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management

International Journal of Recent Contributions from Engineering, Science & IT (iJES), 2015

For the past five years, the bitcoin network constantly experience a growth in its size as more c... more For the past five years, the bitcoin network constantly experience a growth in its size as more communities turn to accept the currency for payment exchanges. Using Flat File and a LevelDB of indices to save blocks on disk, bitcoin users require more memory to save the history of transaction. We focus on issues of memory management and access time in the bitcoin protocol using SQLite DataBase. With all the advantages of SQLite DataBase, it would be efficient if it is fitted in this architecture. The SQLite comes with many flavors one of which is its ability to support sql queries. Thus, instead of parsing indices to search a block from the database, a more powerful query can do the job.

Proceedings of the 4th International Conference on Theory and Practice of Electronic Governance - ICEGOV '10, 2010

In this paper, we propose VeSMDiag -- a Very Short Message Protocol based system for Medical Diag... more In this paper, we propose VeSMDiag -- a Very Short Message Protocol based system for Medical Diagnosis. VeSMDiag is a tool for enabling remote access to the medical diagnosis of first level through mobile phones.