A5086805519- Topic Modeling
- Advanced Text Analysis Techniques
- Natural Language Processing Techniques
- Sentiment Analysis and Opinion Mining
- Semantic Web and Ontologies
- Text and Document Classification Technologies
- Web Data Mining and Analysis
- Multi-Agent Systems and Negotiation
- Peer-to-Peer Network Technologies
- Spam and Phishing Detection
- Advanced Database Systems and Queries
- Speech and dialogue systems
- Biomedical Text Mining and Ontologies
- Authorship Attribution and Profiling
- Recommender Systems and Techniques
- Service-Oriented Architecture and Web Services
- AI-based Problem Solving and Planning
- Information Retrieval and Search Behavior
- Complex Network Analysis Techniques
- Misinformation and Its Impacts
- Asian Culture and Media Studies
- Data Quality and Management
- Smart Agriculture and AI
- Logic, Reasoning, and Knowledge
- Customer Service Quality and Loyalty
British University in Dubai
2015-2021
IBM (Egypt)
2020
Nile University
2012-2019
IEEE Computer Society
2015-2019
Nile University
2015-2017
Charles University
2016
Cairo University
2006-2011
Association for Computing Machinery
2009
Ministry of Agriculture and Land Reclamation
2004
Agricultural Research Center
2004
Advancements in neural networks have led to developments fields like computer vision, speech recognition and natural language processing (NLP). One of the most influential recent NLP is use word embeddings, where words are represented as vectors a continuous space, capturing many syntactic semantic relations among them. AraVec pre-trained distributed representation (word embedding) open source project which aims provide Arabic research community with free powerful embedding models. The first...
With the rapid increase in volume of Arabic opinionated posts on different microblogging mediums comes an increasing demand for sentiment analysis tools. Yet, research area is progressing at a very slow pace compared to that being carried out English and other languages. This paper highlights major problems open issues face social media. The also presents case study goal which investigate possibility determining semantic orientation Egyptian tweets comments given limited resources. One...
This paper presents the results and conclusions of our participation in SemEval-2017 task 8: Determining rumour veracity support for rumours.We have participated 2 subtasks: SDQC (Subtask A) which deals with tracking how tweets orient to accuracy a rumourous story, Veracity Prediction B) goal predicting given rumour.Our was closed variant, prediction is made solely from tweet itself.For subtask A, linear vector classification applied model bag words, help naïve Bayes classifier used semantic...
research-article Share on A panoramic survey of natural language processing in the Arab world Authors: Kareem Darwish Hamad Bin Khalifa University, Doha, Qatar QatarView Profile , Nizar Habash New York University Abu Dhabi, United Emirates EmiratesView Mourad Abbas (CRSTDLA), Bouzareah, Algeria AlgeriaView Hend Al-Khalifa King Saud Riyadh, Saudi Arabia ArabiaView Huseein T. Al-Natsheh Mawdoo3, Jordan JordanView Houda Bouamor Carnegie Mellon Karim Bouzoubaa Mohammed V Rabat, Morocco...
The problem that ontology learning deals with is the knowledge acquisition bottleneck, to say difficulty actually model relevant domain of interest.Ontologies are vehicle by which we can and share among various applications in a specific domain.So many research developed several approaches systems.In this paper, present survey for different from semi-structured unstructured date
This paper describes two systems that were used by the NileTMRG for addressing Arabic Sentiment Analysis as part of SemEval-2017, task 4. participated in three related subtasks which are: Subtask A (Message Polarity Classification), B (Topic-Based Message classification) and D (Tweet quantification). For subtask A, we made use NU's sentiment analyzer augmented with a scored lexicon. D, an ensemble different classifiers. The first classifier was convolutional neural network trained (word2vec)...
This paper explores the impact of taking environment within which a tweet is made, on task analyzing sentiment orientations tweets produced by people in same community. The proposes C-GRU (Context-aware Gated Recurrent Units), extracts contextual information (topics) from and uses them as an extra layer to determine sentiments conveyed tweet. proposed architecture learns direct co-relations between such task's predication. multi-modal model combines both outputs learnt (from topics...
Ontologies play a vital role in many web- and internet-related applications. This work presents system for accelerating the ontology building process via semi-automatically learning hierarchal given set of domain-specific web documents seed concepts. The methods are tested with domain agriculture. is constructed through use two complementary approaches. presented has been used to build an agricultural using Arabic extension evaluated against modified version AGROVOC ontology.
Stemming is a key step in most text mining and information retrieval applications. Information extraction, semantic annotation, as well ontology learning are but few examples where using stemmer must. While the use of light stemmers Arabic texts has proven highly effective for task retrieval, this class falls short providing accuracy required by many This can be attributed to fact that employ set rules they apply indiscriminately do not address stemming broken plurals at all, even though...
This paper describes our participation in the SemEval-2016 task 5, Aspect Based Sentiment Analysis (ABSA).We participated two slots sentence level ABSA (Subtask 1) namely: aspect category extraction (Slot and sentiment polarity 3) English Restaurants Laptops reviews.For Slot 1, we applied different models for each domain.In restaurants domain, used an ensemble classifier which is a combination of Convolutional Neural Network (CNN) initialized with pretrained word vectors, Support Vector...
This paper explores the idea of dynamically adding multi-destination links to Web pages, based on context pages and users, as a way assisting users in their information finding navigation activities. The work does not make any preconceived assumptions about needs its users. Instead it presents method for generating by adapting community utilizing these within this individual needs. implementation is carried out multi-agent framework where concepts from open hypermedia are extended exploited....
Research addressing Sentiment Analysis has witnessed great attention over the last decade especially after huge increase in social media networks usage. Social like Facebook and Twitter generate an incredible amount of data on a daily basis, containing posts that discuss all kinds different topics ranging from sports products to politics current events. Since generated within these mediums is created by users world, it multilingual nature. Arabic one important languages recently targeted...
Arabic Twitter Sentiment Analysis has been gaining a lot of attention lately with supervised approaches being exploited widely. However, to date, there not an experimental study that examines how different configurations the Bag Words model, text representation scheme, can affect various machine learning methods. The goal presented work is do exactly that. Specifically, this which best for each three have shown good results when applied on task sentiment analysis, namely: Support Vector...
This work exploits the logical structure of information rich texts to automatically annotate text segments contained within them using a domain ontology. The underlying assumption behind this is that in such documents embody self informative units. Another segment headings coupled with document's hierarchical offer informal representations content; and matching concepts an ontology/thesaurus can result creation formal labels/meta-data for these segments. When encountered heading not be...
Automatic key phrase extraction has many important applications including but not limited to summarization, cataloging/indexing, feature for clustering and classification, data mining. This paper presents a simple, yet effective algorithm (KP-Miner) achieving this task. The result of an experiment carried out investigate the effectiveness is also presented. In devised applied six different datasets consisting 481 documents. results are then compared two existing sophisticated machine...