Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userText WaterMarking
إخفاء المعلومات في النصوص
780
1 0
0.0
(
0
)
Added by
Syrian Virtual University رسالة ماجستير
Publication date
2017
fields
Informatics Engineering
and research's language is
العربية
Created by
Shamra Editor
Ask ChatGPT about the research
مع تطور الشبكات الحاسوبية لوحظ تزايد النشر الالكتروني للمنشورات عن طريق قواعد البيانات المتوفرة على هذه الشبكات CD-ROM بالإضافة إلى المكتبات الالكترونية
No English abstract
References used
Jeng-Shyang Pan, Hsiang- Cheh Huang L. C. Jain . Intelligent Watermarking Techniques, Part 4. ISBN 981-238-955-5
rate research
Read More
Representations from large pretrained models such as BERT encode a range of features into monolithic vectors, affording strong predictive accuracy across a range of downstream tasks. In this paper we explore whether it is possible to learn disentangl
ed representations by identifying existing subnetworks within pretrained models that encode distinct, complementary aspects. Concretely, we learn binary masks over transformer weights or hidden units to uncover subsets of features that correlate with a specific factor of variation; this eliminates the need to train a disentangled model from scratch for a particular task. We evaluate this method with respect to its ability to disentangle representations of sentiment from genre in movie reviews, toxicity from dialect in Tweets, and syntax from semantics. By combining masking with magnitude pruning we find that we can identify sparse subnetworks within BERT that strongly encode particular aspects (e.g., semantics) while only weakly encoding others (e.g., syntax). Moreover, despite only learning masks, disentanglement-via-masking performs as well as --- and often better than ---previously proposed methods based on variational autoencoders and adversarial training.
In this paper, a technique for hiding a message into digital images
(256 gray levels) is proposed. Some tests, on the proposed technique, are
done by applying the technique in hiding messages into some images. From
the recorded results, we can not
e that the proposed watermarking technique
introduces, in many cases, a reasonable way that can be used to hide secret
messages in the digital images.
تسببت الزيادة الكبيرة في كمية المعلومات المتاحة في الانترنت من مختلف المصادر في السنوات الأخيرة إلى صعوبة الوصول والبحث في النصوص الكبيرة عن المعلومة المطلوبة بسرعة وكفاءة وكان من الصعب جداً استخراج تلاخيص النصوص بشكل يدوي وذلك بسبب النمو الهائل للمع
لومات بشكل يومي لذلك أصبح من الضروري استخراج التلاخيص تلقائياً من نص واحد أو عدة نصوص لذلك سنتطرق في بحثنا إلى أهم الأساليب والطرق في عمليات التلخيص في الأعوام السابق
يهدف التنقيب في النصوص بشكل عام إلى تحليل النصوص لاستخلاص معارف ذات جودة عالية من عدة مصادر نصية، والربط فيما بينها لتشكيل حقائق وفرضيات جديدة. تعد الأوراق البحثية التمثيل الأكثر اكتمالاً للمعرفة البشرية. وقد ساهمت حركة "الوصول المفتوح" إلى الأوراق ا
لبحثية، بالإضافة إلى ازدهار حقل التعلم الآلي في الآونة الأخيرة وتوفر الأدوات البرمجية والعتادية بكلف منخفضة نسبياً، بتداعي الحواجز المعيقة لعملية التنقيب في نصوص الأوراق البحثية.
في تتمة هذه الدراسة سنستعرض مجموعة من أساليب التنقيب في النصوص العلمية من حيث أهميتها، مجالات استخدامها، وطرق تطبيقها.
Translation is one of the oldest types of human activity, as it is a phenomenon inherent in human history from the oldest ages, as it appeared with the emergence of the need for a means of understanding between the peoples of the world due to the dif
ference in the management of communication between them.
Translation, like other fields of knowledge, was subject to the same rule of tradition mentioned earlier, so that when you look at the books of some Arab thinkers, you will find that the nature of their writings is of a foreign nature.
Among the aims to be achieved in this research are:
Introducing the theory of text types and their relationship to translation.
Disclosure of the characteristics of this theory in translation.
Clarify the importance of text types theory and its role in acquiring translation skills.
Linguistic analysis of texts as a first station, followed by other consistent and harmonious stations that collectively work to achieve the communicative relationship of the various types of texts.
Identifying the types of texts and then the types of translation approaches that work to reduce the gap between the source text and the access text.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
