ARABIC SCRIPT LANGUAGE IDENTIFICATIONS USING ADAPTIVE NEURAL NETWORK Ali Selamat and Ng Choon-Ching Universiti Technologi Malaysia Faculty of Computer Science and Information Systems
[email protected],
[email protected] ABSTRACT Globalization has led to increase in information flows between geographically remote locations and realization of a global common market. When building website applications for use on various industries, developers need to deal with a wide range of users from different countries. Thus, multilingual system is implemented in order to provide multilingual environment in those applications. However, it is time-consuming to define all the possible languages for multilingual system manually, it would be desirable to automate the adaption of language identification for text-based documents. To address this need, we introduce language identification of Arabic script documents with letter frequency based. Techniques used for identification are fuzzy ARTMAP and default ARTMAP, which are belong to neural network architectures that perform incremental supervised learning. Arabic script documents such as Arabic, Persian and Urdu were used for performing language identification. From the experiments, we have found that fuzzy ARTMAP has performed better than the default ARTMAP in Arabic script language identification. KEY WORDS Adaptive neural networks, Fuzzy ARTMAP, Default ARTMAP, letter frequency, language identification, Arabic script.
1 Introduction Language identification is the process of recognizing the identity language of human communication for the given content. For instances, languages such as English, German, French, Croatian, Mandarin, Japanese, Arabic etc. Without the basic knowledge for the given content is written in, applications such as information retrieval and text mining are not able to accurately process the data, potentially leading to a loss of critical information. The problem of text-based language identification is attempted to be solved for long time and various feature based models were experimented for written language identification. Nowadays, there is a lot of information across the internet either in text or image format which are encoded in different languages. Many information are also coded by a string of characters such as DNA representation (CGTA), web page syntaxes (,
, <body>), short form (before<br />
<br />
- b4, what - wat) etc. Consequently, those noisy data will affect features selection from document when performing language identification. Furthermore, unstructured words, sentences, or paragraphs need to be taken into consideration when preprocessing data. Many methods have been explored for language identification of textual document such as neural network [1, 2, 3, 4], n-gram based algorithm [5, 6, 7, 8], vector space modeling [9] etc. During the past decades, many research works have been done in the area of natural language identification to keep up with the growing demands set by the multilingual systems. Consequently, some of the works have been successfully developed into products which are freely downloaded or available only when purchased. For examples, van Noord’s TextCat [10], Basic Tech’s Rosetta Language Identifier [11], and web based language identification services such as Xerox Language Identifier [12]. Adaptive Resonance Theory (ART) is a family of algorithms for unsupervised learning developed by Carpenter and Grossberg [13, 14, 15]. ART is similar to many iterative clustering algorithms where each sample is processed by finding the ”nearest” cluster and then updating that cluster to be ”closer” to the example. ART allows a training example to modify an existing cluster only if the cluster is sufficiently close to the example (the cluster is said to ”resonate” with the example). Otherwise, a new cluster is created to handle the example. In addition, ART uses a vigilance parameter as a threshold of similarity between patterns and clusters to determine whether a new cluster should be formed. ART was motivated by the ”stability-plasticity dilemma”, a term used by Grossberg that describes the problems belonging exclusively to competitive learning. ART resolves this problem by creating a new cluster every time an example is very dissimilar from the existing cluster. However, in the presence of noisy data, ART has a tendency to create new clusters continuously, resulting in ”category proliferation” or ”overfitting” [16, 17, 18]. Unlike most off-line neural network models, ARTbased networks, including Fuzzy ARTMAP (FAM) [19] and Default ARTMAP (DAM) [20], have a ”growing” architecture. FAM and DAM makes use of a clustering algorithm to bring into relation of input pattern with their outputs. Basically, FAM is an extension of the ARTMAP network [21]. It replaces the operation of conventional set the-<br />
<br />
ory in an ARTMAP network with fuzzy set theory. DAM is resulted from the comparative analysis of various basic network such as FAM, ART-EMAP [22], ARTMAP-IC [23], Gaussian ARTMAP [24], and distributed ARTMAP [13, 25]. In this paper, we are focusing on the identification of Arabic script languages using cognitive neural network, FAM and DAM. We analyzed the letter frequency based method in selecting the best features for language identification. Language Predicted<br />
<br />
Output Class Prediction<br />
<br />
category proliferation when noisy inputs are trained with fast learning. Distributed activation improves noise tolerance and the code compression while new system dynamics retain the stable fast learning capabilities of winner-take-all ART systems. For winner-take-all scheme [20], the model will search for a chosen coding node (WTA) J that meets the matching criterion and predicts the correct output class K, as follows: Step 1. Code: for the next sorted coding node (j=J) that meets the matching criterion, ¸ · |A ∧ WJ | ≥ ρ, set yJ = 1 (W T A) (1) M Step 2. Output class prediction,<br />
<br />
Training<br />
<br />
σk =<br />
<br />
Testing<br />
<br />
C X<br />
<br />
Wjk yj =WJk<br />
<br />
(2)<br />
<br />
j=1<br />
<br />
A is the complement coded input vector, WJ is the output class weight vector, M is number of input features, ρ is the vigilance variable, yJ is the coding field activation pattern (CAM), C is the number of committed coding nodes, k is the output class index, j is the coding node index and J is the chosen coding node (winner-take-all).<br />
<br />
A (complement coded input vector)<br />
<br />
input patterns<br />
<br />
Method<br />
<br />
FAM<br />
<br />
Training Testing<br />
<br />
Winner - Take -All Winner - Take -All<br />
<br />
DAM Winner -Take -All Distributed Category Activation Scheme<br />
<br />
Figure 1. Overview of ARTMAP architecture<br />
<br />
A = (a, ac )<br />
<br />
and<br />
<br />
|A| = M<br />
<br />
(3)<br />
<br />
For distributed category activation scheme [20], increased gradient (IG) CAM rule is implemented for output class prediction as follow: Step 1: Point box case: If Λ0 6= φ (i.e., wj = A for some j), set yj = |Λ10 | for each j ∈ Λ0<br />
<br />
2<br />
<br />
Methods<br />
<br />
ARTMAP is preferred due to its incremental learning ability. As new data is sampled, there will be no need to retrain the network as would be the case with the multi layer perceptron of neural network. Figure 1 shows the ARTMAP architecture and different between FAM and DAM. Input patterns from data preprocessing will be accepted in ARTMAP as complement coded input vector, A. The selected input vector will be used for ARTMAP training and the remaining input vector will be used for ARTMAP testing. Finally, the output class prediction of ARTMAP will be justified whether meets the matching criterion. In other words, the ARTMAP will update the weight if the input found the similar category, else it will create a new category for that input. FAM uses winner-take-all (WTA) scheme during training and testing. However, DAM uses winner-take-all scheme during training and distributed category activation scheme during testing. In ARTMAP networks, winner-take-all competitive activation supports stable coding, but this limiting case of competition may cause<br />
<br />
Step 2: If Λ0 = φ, set h<br />
<br />
1 M −Tj<br />
<br />
yj = P h λ∈Λ<br />
<br />
ip<br />
<br />
1 M −Tλ<br />
<br />
ip<br />
<br />
f or each j ∈ Λ<br />
<br />
(4)<br />
<br />
Step 3. Output class prediction, σk =<br />
<br />
C X<br />
<br />
Wjk yj =WJk<br />
<br />
(5)<br />
<br />
j=1<br />
<br />
Λ is the component-wise minimum and Tj is the signal from input field to coding node j. 2.1<br />
<br />
Fuzzy ARTMAP<br />
<br />
The fuzzy ARTMAP architecture is an auto-organized learning system. This kind of network has supervised training and pertains to the Adaptive Resonance Theory (ART)<br />
<br />
family; its structure is based on the adaptive resonance theory and is similar to the fuzzy ART network, which employs calculus based on fuzzy logic. The inter-ART module has a self-regulator mechanism named match tracking, whose objective is to maximize the generalization and minimize the network error [19]. The fuzzy ARTMAP operates by dividing the input space into a number of clusters, which are mapped into an output space. Instance based learning is used, where each sample is mapped to a class label. Three parameters namely the vigilance ρ ∈ [0 ∼ 1], the learning rate β ∈ [0 ∼ 1], and the choice parameter α, are used to control the learning process. The choice parameter is generally made small and a value of 0.01 was used in this work. The parameter α controls the adaption speed, where 0 implies a slow speed and 1, the fastest. If α = 1, the clusters get enlarged to include the point represented by the input vector. The vigilance represents the degree of belonging and it controls how large any cluster can become, resulting in new clusters being formed. Larger values of ρ lead to a case where smaller cluster are formed and this eventually lead to ”category proliferation”, which can be viewed as overtraining. In this work, different vigilance parameters were experimented to validate the accurately identification on Arabic script.<br />
<br />
a free parameter, five voters have proven to be sufficient for many applications. Default ARTMAP thus trains five voting networks for each training set combination [20].<br />
<br />
3<br />
<br />
Data Collection<br />
<br />
Dataset were randomly collected from British Broadcasting Corporation (BBC) website manually. Those dataset were save in UTF-8 format by setting the file name correspond to its languages. For instances, text collected from selected Arabic BBC news will save as ”a1.txt” (a1 means first Arabic document). This process is repeated till two hundreds documents had been collected for each language. Consequently, tagged UTF-8 documents were prepared for evaluation. Figure 2 shows the processes of data collection in the initial part (1 and 2). Preprocessing stage is excluded due to the fact that this work is based on letter frequency.<br />
<br />
BBC Website (Arabic, Persian and Urdu) 1<br />
<br />
Tagged UTF-8 Documents 2<br />
<br />
2.2 Default ARTMAP The character of their code representations, distributed or winner-take-all, is a primary factor differentiating various ARTMAP networks. The original models [21, 19] employ winner-take-all coding during training and testing, as do many subsequent variations and the majority of ART systems that have been transferred to technology. Default ARTMAP codes a training input as a winner-take-all activation pattern, but codes a test input as a distributed activation pattern. For distributed coding, the transformation of the filtered bottom-up input to an activation pattern across a field of nodes is defined by the increased-gradient CAM rule [25]. The default network also implements the MTsearch algorithm [23] and sets the baseline vigilance parameter ρ¯ equal to zero, for optimum code compression. Other design choices for default ARTMAP include fast learning, whereby weights converge to asymptote on each training trial; single-epoch training, which emulates on-line learning; a choice-by-difference signal function from the input field to the coding field; and four-fold cross validation. ARTMAP’s capacity for fast learning implies that the system can incorporate information from examples that are important but infrequent and can be trained incrementally. Fast learning also causes each network’s memory to vary with the order of input presentation during training. Voting across several networks trained with different orderings of a given input set takes advantage of this feature, typically improving performance and reducing variability as well as providing a measure of confidence in each prediction. While the number of voting systems is, in general,<br />
<br />
Feature Extraction (Letter Frequency)<br />
<br />
Unicode Code Point Convertion 4<br />
<br />
3<br />
<br />
Normalization (scaled between 0 - 1)<br />
<br />
Input Patterns 5<br />
<br />
6<br />
<br />
8<br />
<br />
7<br />
<br />
Language Identified<br />
<br />
Classifier (FAM and DAM)<br />
<br />
Figure 2. Flow of language identification<br />
<br />
3.1<br />
<br />
Letter Frequency<br />
<br />
Before processing the data collected, we used the unicode code point to convert the original Arabic script letter or character into its corresponding value in decimal (view Table 1). Next, we find out the significant letters frequently exist in each language. For each document, only the first 200 letters were evaluated for the frequency of occurrence. If the letter found is noisy data (excluded from Arabic script), it will be continued to next letter. Otherwise, the matched<br />
<br />
Table 1. Examples of corresponding unicode code point in decimal<br />
<br />
H<br />
<br />
Letter Code Point<br />
<br />
H<br />
<br />
1578<br />
<br />
1579<br />
<br />
h. 1580<br />
<br />
h<br />
<br />
(%) 100<br />
<br />
95.36<br />
<br />
94.64<br />
<br />
p<br />
<br />
1581<br />
<br />
Accuracy Model<br />
<br />
92.68 95<br />
<br />
91.43 88.57<br />
<br />
88.04<br />
<br />
1582<br />
<br />
90<br />
<br />
87.32<br />
<br />
84.64<br />
<br />
85<br />
<br />
letter frequency will be added one each time. However, some additional letters of each language are also excluded from evaluation (view Table 2). Those letters cannot be considered as standard letter found in Arabic script languages. Result will show in the following section. After that, 5 top frequency letters of each Arabic script language will be selected as features of input patterns. Normalization will take place to scale the data between 0 and 1. Finally, input patterns are ready to be processed by classifier. Table 2. Excluded Arabic script letters<br />
<br />
P<br />
<br />
h<br />
<br />
H<br />
<br />
H<br />
<br />
X<br />
<br />
P<br />
<br />
Letter<br />
<br />
À<br />
<br />
Code Point<br />
<br />
1711 1688 1670 1662 1657 1672 1681<br />
<br />
DAM FAM<br />
<br />
80 75 70 1 st 200<br />
<br />
1 st 300<br />
<br />
1<br />
<br />
st<br />
<br />
400<br />
<br />
1<br />
<br />
st<br />
<br />
500<br />
<br />
Total letter used<br />
<br />
Figure 3. The accuracy network for different number of letters used<br />
<br />
is a encouraging result since there are limitless data need to be processed in real-time. Next, we used every first 500 letters as experiment benchmark. 98 96 94<br />
<br />
Table 3. Partial results of frequency letter evaluation Letter Arabic Persian Urdu Legend Letter Arabic Persian Urdu<br />
<br />
4<br />
<br />
@ <br />
<br />
@<br />
<br />
ð <br />
<br />
@<br />
<br />
ø<br />
<br />
324 2732 1414<br />
<br />
5173 59 0<br />
<br />
416 16 244<br />
<br />
2294 0 0<br />
<br />
1448 580 3540<br />
<br />
- 176 Arabic script letters - collected 200 Arabic Documents - collected 200 Persian Documents - collected 200 Urdu Documents<br />
<br />
Experimental Results and Discussion<br />
<br />
First of all, each 200 documents of Persian, Arabic and Urdu language were evaluated on frequency letter. Table 3 shows the partial result from the evaluation. We observed that 15 significant letters which are representing those lan guages in Arabic script. There are @, @, è, H , ¼, @, X, P, , è,<br />
<br />
à, ë, è, þ, - . As a result, the following experiments used<br />
<br />
those letters as features of each document. Figure 3 shows the accuracy network for different number of letters used in 600 Arabic script documents. This experiment used every first 200-500 letters from each document. We observed that if the number of letters used is increasing, the performance of network getting better. It<br />
<br />
AccuracyNetwork(%)<br />
<br />
92<br />
<br />
DAM<br />
<br />
90<br />
<br />
FAM<br />
<br />
88 86 84 82 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Baseline vigilance (rho)<br />
<br />
Figure 4. The accuracy network for different vigilance parameters Vigilance parameter is highly influence on code compression. Figure 4 shows the experiment on various vigilance parameters (rho = 0.1 ∼ 0.9). Both algorithm shown highest accuracy network (FAM : 96.61%, DAM : 94.29%) on rho = 0.8. However, we noticed that higher values of vigilance parameter increase accuracy network but may decrease code compression. In the last experiment, we duplicate the existing documents by randomly mixed context of two documents into one document. For example, context of a99.txt and a100.txt are combined together by randomizing it and formed a new document, a5000.txt. This is only for simulation purpose in order to justify the feasibility by using letter frequency based features in language identification of Arabic script. Figure 5 shows that accuracy network performed better when increasing the number of documents experimented. This achievement indicate that letter frequency based is very useful in text-based language identification. Besides,<br />
<br />
it is also very applicable with both fuzzy ARTMAP and default ARTMAP. 100 95<br />
<br />
AccuracyNetwork(%)<br />
<br />
90 85 80 75 70 50<br />
<br />
100<br />
<br />
200<br />
<br />
500<br />
<br />
1000<br />
<br />
5000<br />
<br />
10000<br />
<br />
Total Documents Used Legend FAM DAM Highest accuracy of FAM, 97.13% Highest accuracy of DAM, 92.85%<br />
<br />
Figure 5. The accuracy network for various number of documents<br />
<br />
In overall, we found that fuzzy ARTMAP is performed better than default ARTMAP. In other words, ARTMAP performed better if using winner-take-all scheme in both training and testing section.<br />
<br />
5 Conclusions and Future Work In this paper, we presented the Arabic script language identification using adaptive neural network, fuzzy ARTMAP (FAM) and default ARTMAP (DAM). The languages involved in Arabic script were Arabic, Persian and Urdu. We used letter frequency as features of input patterns for the network. Our experiments show that the letter frequency based approach are feasible and extensible , and can be processed smoothly with adaptive neural network. Higher vigilance parameter increases the accuracy of identification but time consuming. Fuzzy ARTMAP is performed better than default ARTMAP. This work will be extended to test on other languages such as English, Malay, German, Japanese, Korea etc. In addition, various feature selection approaches will be implemented on text-based documents for optimizing the accuracy of language identification.<br />
<br />
5.1 Acknowledgement This work is supported by the Ministry of Science &Technology and Innovation (MOSTI), Malaysia and Research Management Center, Universiti Teknologi Malaysia (UTM) under the Vot 79089.<br />
<br />
References [1] E.B. Bilcu and J. Astola, A hybrid network for language identification from text, Proceedings of the 16th IEEE Signal Processing Society Workshop on Machine Learning for Signal Processing, 2006, 253-258. [2] E.B. Bilcu, J. Astola and J. Saarinen, A hybrid network / rule based system for bilingual text-to-phoneme mapping, Proceedings of the 14th IEEE Signal Processing Society Workshop on Machine Learning for Signal Processing, 2004, 345354. [3] J. Tian and J. Suontausta, Scalable neural network based language identification from written text, Proceedings of the 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 1, 2003, 48-51. [4] M.J. Embrechts and F. Arciniegas, Neural network for textto-speech phoneme recognition, Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics, 5, 2000, 3582-3587. [5] N. Ljubesic, N. Mikelic and D. Boras, Language identification: How to distinguish similar languages?, Proceedings of the 29th International Conference on Information Technology Interfaces, 2007, 541-546. [6] L.F. Zhai, M.H. Siu, X. Yang and H. Gish, Discriminatively trained language models using support vector machines for language identification, Proceedings of the IEEE Odyssey on Speaker and Language Recognition Workshop, 2006, 1-6. [7] B. Martins and M.J. Silva, Language identification in web pages, Proceedings of the 2005 ACM symposium on Applied Computing, 2005, 764-768. [8] J. Hakkinen and J. Tian, N-gram and decision tree based language identification for written words, Proceedings of the IEEE Workshop on Automatic Speech Recognition and Understanding, 2001, 335-338. [9] H.Z. Li, B. Ma and C.H. Lee, A vector space modeling approach to spoken language identification, IEEE Transactions on Audio, Speech, and Language Processing, 15(1), 2007, 271-284. [10] TextCat Language Guesser Demo, available from ac<http://www.let.rug.nl/vannoord/TextCat/Demo>, cessed on [01/06/2007] [11] Basis Techis Rosetta Language Identifier, available from <http://www.basistech.com/language-identification>, accessed on [01/06/2007] [12] Xeror Language Identifier, available from <http://www.xrce.xerox.com/research.html>, accessed on [01/06/2007] [13] G.A. Carpenter, Distributed learning, recognition, and prediction by ART and ARTMAP neural networks, Neural Networks, 10, 1997, 1473-1494. [14] S. Grossberg, The link between brain, learning, attention, and consciousness, Consciousness and Cognition, 8, 1999, 1-44. [15] S. Grossberg, How does the cerebral cortex work? Development, learning, attention, and 3D vision by laminar circuits of visual cortex, Behavioral and Cognitive Neoroscience Reviews, 2, 2003, 47-76. [16] G.A. Carpenter and S. Grossberg, A massively parallel architecture for a self-organizing of stable category recognition codes for analog input patterns, Computer Vision, Graphics, and Image Processing, 37(1), Jan. 1987a, 54-115. [17] G.A. Carpenter and S. Grossberg, ART 2: Self-organization of stable category recognition codes for analog input patterns, Applied Optics, Dec. 1987b, 4919-4930.<br />
<br />
[18] G.A. Carpenter and S. Grossberg, The ART of adaptive pattern recognition by a self-organizing neural network, Computer, 21(3), March 1988, 77-88. [19] G.A. Carpenter, S. Grossberg, N. Markuzon, J.H. Reynolds, and D.B. Rosen, Fuzzy ARTMAP: A neural network architecture for incremental supervised learning for analog multidimensional maps, IEEE Transactions on Neural Networks, 3, 1992, 698-713. [20] G.A. Carpenter, Default ARTMAP, Proceedings of the International Joint Conference on Neural Networks (IJCNN’03), 2, 2003, 1396-1401. [21] G.A. Carpenter, S. Grossberg and J.H. Reynolds, ARTMAP: Supervised real-time learning and classification of nonstationary data by a self-organizing neural network, Neural Networks, 4(5), 1991, 565-588. [22] G.A. Carpenter and W.D. Ross, ART-EMAP: A neural network architecture for object recognition by evidence accumulation, IEEE Transactions on Neural Networks, 6, 1995, 805-818. [23] G.A. Carpenter and N. Markuzon, ARTMAP-IC and medical diagnosis: Instance counting and inconsistent cases, Neural Networks, 11, 1998, 323-336. [24] J.R. Williamson, Gaussian ARTMAP: A neural network for fast incremental learning of noisy multidimensional maps, Neural Networks, 9, 1998, 881-897. [25] G.A. Carpenter, S. Grossberg and B.W. Noeske, Distributed ARTMAP: A neural network for fast distributed supervised learning, Neural Networks, 11, 1998, 793-813.<br />
<br />
</div>
</div>
</div>
</div>
</div>
</div>
<div class="row hidden-xs">
<div class="col-md-12">
<h4></h4>
<hr />
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/certificate-in-arabic-language_5b1668097f8b9a317a8b456e.html">
<img src="https://p.pdfkul.com/img/300x300/certificate-in-arabic-language_5b1668097f8b9a317a8b456e.jpg" alt="CERTIFICATE IN ARABIC LANGUAGE" height="200" class="block" />
<h4 class="name-title">CERTIFICATE IN ARABIC LANGUAGE</h4>
</a>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/certificate-in-arabic-language_59bdb04b1723dd98e83397e8.html">
<img src="https://p.pdfkul.com/img/300x300/certificate-in-arabic-language_59bdb04b1723dd98e83397e8.jpg" alt="CERTIFICATE IN ARABIC LANGUAGE" height="200" class="block" />
<h4 class="name-title">CERTIFICATE IN ARABIC LANGUAGE</h4>
</a>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/arabic-language-resources-and-tools_59be0df71723dd98e8339d4b.html">
<img src="https://p.pdfkul.com/img/300x300/arabic-language-resources-and-tools_59be0df71723dd98e8339d4b.jpg" alt="Arabic Language Resources and Tools" height="200" class="block" />
<h4 class="name-title">Arabic Language Resources and Tools</h4>
</a>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/direct-adaptive-control-using-an-adaptive-reference-model_59d162161723dd0732b93154.html">
<img src="https://p.pdfkul.com/img/300x300/direct-adaptive-control-using-an-adaptive-referenc_59d162161723dd0732b93154.jpg" alt="Direct adaptive control using an adaptive reference model" height="200" class="block" />
<h4 class="name-title">Direct adaptive control using an adaptive reference model</h4>
</a>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/direct-adaptive-control-using-single-network-adaptive-_5a0b626f1723dd870ff5e168.html">
<img src="https://p.pdfkul.com/img/300x300/direct-adaptive-control-using-single-network-adapt_5a0b626f1723dd870ff5e168.jpg" alt="Direct Adaptive Control using Single Network Adaptive ..." height="200" class="block" />
<h4 class="name-title">Direct Adaptive Control using Single Network Adaptive ...</h4>
</a>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/rule-based-approach-in-arabic-natural-language-_59bb50111723dde2a958cb68.html">
<img src="https://p.pdfkul.com/img/300x300/rule-based-approach-in-arabic-natural-language-_59bb50111723dde2a958cb68.jpg" alt="Rule-based Approach in Arabic Natural Language ..." height="200" class="block" />
<h4 class="name-title">Rule-based Approach in Arabic Natural Language ...</h4>
</a>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/rule-based-approach-in-arabic-natural-language-_5a170bb31723dd269d9d3fb5.html">
<img src="https://p.pdfkul.com/img/300x300/rule-based-approach-in-arabic-natural-language-_5a170bb31723dd269d9d3fb5.jpg" alt="Rule-based Approach in Arabic Natural Language ..." height="200" class="block" />
<h4 class="name-title">Rule-based Approach in Arabic Natural Language ...</h4>
</a>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/any-african-language-german-russian-arabic-services_5acfd4c67f8b9a165a8b4583.html">
<img src="https://p.pdfkul.com/img/300x300/any-african-language-german-russian-arabic-service_5acfd4c67f8b9a165a8b4583.jpg" alt="Any African language German Russian Arabic ... Services" height="200" class="block" />
<h4 class="name-title">Any African language German Russian Arabic ... Services</h4>
</a>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/rule-based-approach-in-arabic-natural-language-_59bb4f211723dde1a9ebca8a.html">
<img src="https://p.pdfkul.com/img/300x300/rule-based-approach-in-arabic-natural-language-_59bb4f211723dde1a9ebca8a.jpg" alt="Rule-based Approach in Arabic Natural Language ..." height="200" class="block" />
<h4 class="name-title">Rule-based Approach in Arabic Natural Language ...</h4>
</a>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/character-identification-in-movie-using-movie-script-ijrit_5a2696361723dd544d438c54.html">
<img src="https://p.pdfkul.com/img/300x300/character-identification-in-movie-using-movie-scri_5a2696361723dd544d438c54.jpg" alt="Character Identification in Movie Using Movie Script - IJRIT" height="200" class="block" />
<h4 class="name-title">Character Identification in Movie Using Movie Script - IJRIT</h4>
</a>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/character-identification-in-movie-using-movie-script-ijrit_5a269aad1723ddf30369dfc0.html">
<img src="https://p.pdfkul.com/img/300x300/character-identification-in-movie-using-movie-scri_5a269aad1723ddf30369dfc0.jpg" alt="Character Identification in Movie Using Movie Script - IJRIT" height="200" class="block" />
<h4 class="name-title">Character Identification in Movie Using Movie Script - IJRIT</h4>
</a>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/using-xdb-to-offload-oracle-reporting-database-to-postgres-script-_59c442141723dd74c9bfbc63.html">
<img src="https://p.pdfkul.com/img/300x300/using-xdb-to-offload-oracle-reporting-database-to-_59c442141723dd74c9bfbc63.jpg" alt="Using xDB to Offload Oracle Reporting Database to Postgres-script ..." height="200" class="block" />
<h4 class="name-title">Using xDB to Offload Oracle Reporting Database to Postgres-script ...</h4>
</a>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/improving-arabic-information-retrieval-system-using-n-gram-method-_59ce64531723dd4a8cd2ed0d.html">
<img src="https://p.pdfkul.com/img/300x300/improving-arabic-information-retrieval-system-usin_59ce64531723dd4a8cd2ed0d.jpg" alt="Improving Arabic Information Retrieval System using n-gram method ..." height="200" class="block" />
<h4 class="name-title">Improving Arabic Information Retrieval System using n-gram method ...</h4>
</a>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4">
<div class="box-product doc">
<div class="doc-meta-thumb name">
<a href="https://p.pdfkul.com/arabic-anaphora-resolution-using-the-web-as-corpus_59c2a4691723dd9b43940fc0.html">
<img src="https://p.pdfkul.com/img/300x300/arabic-anaphora-resolution-using-the-web-as-corpus_59c2a4691723dd9b43940fc0.jpg" alt="Arabic Anaphora Resolution Using the Web as Corpus" height="200" class="block" />
<h4 class="name-title">Arabic Anaphora Resolution Using the Web as Corpus</h4>
</a>
</div>
</div>
</div>
</div>
</div>
<div class="col-lg-3 col-md-4 col-xs-12">
<div class="panel-meta panel panel-info">
<div class="panel-heading">
<h2 class="text-center panel-title">arabic script language identifications using adaptive ...</h2>
</div>
<div class="panel-body">
<div class="row">
<div class="col-md-12">
<span class="st">there is a lot of information across the <em>internet</em> either in text or image format ... works have been done in the <em>area</em> of natural language iden- tification to keep up<wbr> ...</wbr></span> </div>
<div class="col-md-12">
<div class="doc">
<hr />
<div class="download-button" style="margin-right: 3px; margin-bottom: 6px;">
<a href="https://p.pdfkul.com/download/arabic-script-language-identifications-using-adaptive-_5a02405f1723ddd52c99261b.html" class="btn btn-success btn-block"><i class="fa fa-cloud-download"></i> Download PDF </a>
</div>
<div class="share-box pull-left" style="margin-right: 3px;">
<!-- Facebook -->
<a href="http://www.facebook.com/sharer.php?u=https://p.pdfkul.com/arabic-script-language-identifications-using-adaptive-_5a02405f1723ddd52c99261b.html" target="_blank" class="btn btn-social-icon btn-facebook">
<i class="fa fa-facebook"></i>
</a>
<!-- Twitter -->
<a href="http://www.linkedin.com/shareArticle?mini=true&url=https://p.pdfkul.com/arabic-script-language-identifications-using-adaptive-_5a02405f1723ddd52c99261b.html" target="_blank" class="btn btn-social-icon btn-twitter">
<i class="fa fa-twitter"></i>
</a>
</div>
<div class="fb-like pull-left" data-href="https://p.pdfkul.com/arabic-script-language-identifications-using-adaptive-_5a02405f1723ddd52c99261b.html" data-layout="button_count" data-action="like" data-size="large" data-show-faces="false" data-share="false"></div>
<div class="clearfix"></div>
<div class="row">
<div class="col-md-12" style="margin-top: 6px;">
<span class="btn pull-left" style="padding-left: 0;"><i class="fa fa-file-pdf-o"></i> 164KB Sizes</span>
<span class="btn pull-left"><i class="fa fa-download"></i> 0 Downloads</span>
<span class="btn pull-left" style="padding-right: 0;"><i class="fa fa-eye"></i> 162 Views</span>
</div>
</div>
<div class="clearfix"></div>
<div class="row">
<div class="col-md-12">
<span class="btn pull-left" style="padding-left: 0;"><a data-toggle="modal" data-target="#report" style="color: #f44336;"><i class="fa fa-handshake-o"></i> Report</a></span>
</div>
</div>
</div>
</div>
</div>
<h4 id="comment"></h4>
<div id="fb-root"></div>
<script>
(function (d, s, id) {
var js, fjs = d.getElementsByTagName(s)[0];
if (d.getElementById(id))
return;
js = d.createElement(s);
js.id = id;
js.src = "//connect.facebook.net/en_GB/sdk.js#xfbml=1&version=v2.9&appId=266776430439748";
fjs.parentNode.insertBefore(js, fjs);
}(document, 'script', 'facebook-jssdk'));
</script>
<div class="fb-comments" data-href="https://p.pdfkul.com/arabic-script-language-identifications-using-adaptive-_5a02405f1723ddd52c99261b.html" data-width="100%" data-numposts="6"></div>
</div>
</div>
<div class="panel-recommend panel panel-success">
<div class="panel-heading">
<h4 class="text-center panel-title">Recommend Documents</h4>
</div>
<div class="panel-body">
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/certificate-in-arabic-language_5b1668097f8b9a317a8b456e.html">
<img src="https://p.pdfkul.com/img/60x80/certificate-in-arabic-language_5b1668097f8b9a317a8b456e.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/certificate-in-arabic-language_5b1668097f8b9a317a8b456e.html">
CERTIFICATE IN ARABIC LANGUAGE </a>
<div class="doc-meta">
<div class="doc-desc">Translate the following Arabic sentences, related to "conversation between father and son, into Arabic. Li. L.4âº: ci.At : (al-abb : hal t'rif ma wajibuna naHwa ...</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/certificate-in-arabic-language_59bdb04b1723dd98e83397e8.html">
<img src="https://p.pdfkul.com/img/60x80/certificate-in-arabic-language_59bdb04b1723dd98e83397e8.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/certificate-in-arabic-language_59bdb04b1723dd98e83397e8.html">
CERTIFICATE IN ARABIC LANGUAGE </a>
<div class="doc-meta">
<div class="doc-desc">CERTIFICATE IN ARABIC LANGUAGE. Term-End ... Apple. 6. Write (10) simple Arabic sentences related to. "conversation inside the vegetable market". 7.</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/arabic-language-resources-and-tools_59be0df71723dd98e8339d4b.html">
<img src="https://p.pdfkul.com/img/60x80/arabic-language-resources-and-tools_59be0df71723dd98e8339d4b.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/arabic-language-resources-and-tools_59be0df71723dd98e8339d4b.html">
Arabic Language Resources and Tools </a>
<div class="doc-meta">
<div class="doc-desc">selection method enhances the performance of Arabic Text Categorization especially for small threshold ... as any application requiring document organization.</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/direct-adaptive-control-using-an-adaptive-reference-model_59d162161723dd0732b93154.html">
<img src="https://p.pdfkul.com/img/60x80/direct-adaptive-control-using-an-adaptive-referenc_59d162161723dd0732b93154.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/direct-adaptive-control-using-an-adaptive-reference-model_59d162161723dd0732b93154.html">
Direct adaptive control using an adaptive reference model </a>
<div class="doc-meta">
<div class="doc-desc">aNASA-Langley Research Center, Mail Stop 308, Hampton, Virginia, USA; ... Direct model reference adaptive control is considered when the plant-model ...</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/direct-adaptive-control-using-single-network-adaptive-_5a0b626f1723dd870ff5e168.html">
<img src="https://p.pdfkul.com/img/60x80/direct-adaptive-control-using-single-network-adapt_5a0b626f1723dd870ff5e168.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/direct-adaptive-control-using-single-network-adaptive-_5a0b626f1723dd870ff5e168.html">
Direct Adaptive Control using Single Network Adaptive ... </a>
<div class="doc-meta">
<div class="doc-desc">in forward direction and hence can be implemented on-line. Adaptive critic based ... network instead of two required in a standard adaptive critic design.</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/rule-based-approach-in-arabic-natural-language-_59bb50111723dde2a958cb68.html">
<img src="https://p.pdfkul.com/img/60x80/rule-based-approach-in-arabic-natural-language-_59bb50111723dde2a958cb68.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/rule-based-approach-in-arabic-natural-language-_59bb50111723dde2a958cb68.html">
Rule-based Approach in Arabic Natural Language ... </a>
<div class="doc-meta">
<div class="doc-desc">structures and entities are neither available nor easily affordable, and 2) for ... Domain rules have ... machine translation, name-entity recognition, and intelligent.</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/rule-based-approach-in-arabic-natural-language-_5a170bb31723dd269d9d3fb5.html">
<img src="https://p.pdfkul.com/img/60x80/rule-based-approach-in-arabic-natural-language-_5a170bb31723dd269d9d3fb5.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/rule-based-approach-in-arabic-natural-language-_5a170bb31723dd269d9d3fb5.html">
Rule-based Approach in Arabic Natural Language ... </a>
<div class="doc-meta">
<div class="doc-desc">based approach in developing their Arabic natural processing tools and systems. ...... at homes and businesses through the Web, Internet and Intranet services.</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/any-african-language-german-russian-arabic-services_5acfd4c67f8b9a165a8b4583.html">
<img src="https://p.pdfkul.com/img/60x80/any-african-language-german-russian-arabic-service_5acfd4c67f8b9a165a8b4583.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/any-african-language-german-russian-arabic-services_5acfd4c67f8b9a165a8b4583.html">
Any African language German Russian Arabic ... Services </a>
<div class="doc-meta">
<div class="doc-desc">google.com/students/emea. Any African language. German. Russian. Arabic. Greek. Serbian. Bosnian. Hebrew. Slovenian. Bulgarian. Hungarian. Spanish. Croatian. Italian. Swedish. Czech. Lathvian. Swiss German. Danish. Lithuanian. Turkish. Dutch. Norwegi</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/rule-based-approach-in-arabic-natural-language-_59bb4f211723dde1a9ebca8a.html">
<img src="https://p.pdfkul.com/img/60x80/rule-based-approach-in-arabic-natural-language-_59bb4f211723dde1a9ebca8a.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/rule-based-approach-in-arabic-natural-language-_59bb4f211723dde1a9ebca8a.html">
Rule-based Approach in Arabic Natural Language ... </a>
<div class="doc-meta">
<div class="doc-desc">structures and entities are neither available nor easily affordable, and 2) for ... Edinburgh, UK (phone: 971-4-3671963; fax: 971-4-3664698; E-mail:.</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/character-identification-in-movie-using-movie-script-ijrit_5a2696361723dd544d438c54.html">
<img src="https://p.pdfkul.com/img/60x80/character-identification-in-movie-using-movie-scri_5a2696361723dd544d438c54.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/character-identification-in-movie-using-movie-script-ijrit_5a2696361723dd544d438c54.html">
Character Identification in Movie Using Movie Script - IJRIT </a>
<div class="doc-meta">
<div class="doc-desc">M.Tech Student, Department of Computer Science and Engineering ... Names for the clusters are then manually selected from the cast list. ..... Video and Image processing in multimedia system, Cloud Computing and Biometric systems.</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/character-identification-in-movie-using-movie-script-ijrit_5a269aad1723ddf30369dfc0.html">
<img src="https://p.pdfkul.com/img/60x80/character-identification-in-movie-using-movie-scri_5a269aad1723ddf30369dfc0.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/character-identification-in-movie-using-movie-script-ijrit_5a269aad1723ddf30369dfc0.html">
Character Identification in Movie Using Movie Script - IJRIT </a>
<div class="doc-meta">
<div class="doc-desc">IJRIT International Journal of Research in Information Technology, Volume 1, Issue ..... In this paper, the preserved statistic properties are utilized and a proposal to .... processing in multimedia system, Cloud Computing and Biometric systems.</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/using-xdb-to-offload-oracle-reporting-database-to-postgres-script-_59c442141723dd74c9bfbc63.html">
<img src="https://p.pdfkul.com/img/60x80/using-xdb-to-offload-oracle-reporting-database-to-_59c442141723dd74c9bfbc63.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/using-xdb-to-offload-oracle-reporting-database-to-postgres-script-_59c442141723dd74c9bfbc63.html">
Using xDB to Offload Oracle Reporting Database to Postgres-script ... </a>
<div class="doc-meta">
<div class="doc-desc">Using xDB to Offload Oracle Reporting Database to Postgres-script-2.pdf. Using xDB to Offload Oracle Reporting Database to Postgres-script-2.pdf. Open.</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/improving-arabic-information-retrieval-system-using-n-gram-method-_59ce64531723dd4a8cd2ed0d.html">
<img src="https://p.pdfkul.com/img/60x80/improving-arabic-information-retrieval-system-usin_59ce64531723dd4a8cd2ed0d.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/improving-arabic-information-retrieval-system-using-n-gram-method-_59ce64531723dd4a8cd2ed0d.html">
Improving Arabic Information Retrieval System using n-gram method ... </a>
<div class="doc-meta">
<div class="doc-desc">Improving Arabic Information Retrieval System using n-gram method.pdf. Improving Arabic Information Retrieval System using n-gram method.pdf. Open. Extract.</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
<div class="row m-0">
<div class="col-md-3 col-xs-3 pl-0 text-center">
<a href="https://p.pdfkul.com/arabic-anaphora-resolution-using-the-web-as-corpus_59c2a4691723dd9b43940fc0.html">
<img src="https://p.pdfkul.com/img/60x80/arabic-anaphora-resolution-using-the-web-as-corpus_59c2a4691723dd9b43940fc0.jpg" alt="" width="100%" />
</a>
</div>
<div class="col-md-9 col-xs-9 p-0">
<a href="https://p.pdfkul.com/arabic-anaphora-resolution-using-the-web-as-corpus_59c2a4691723dd9b43940fc0.html">
Arabic Anaphora Resolution Using the Web as Corpus </a>
<div class="doc-meta">
<div class="doc-desc">The only study of AR in Arabic, to the best of the authors' knowledge, is that of [16], who have studied AR .... play a decisive role in tracking down the antecedent from a set of possible candidates. ... [16] built WSD engine using hit counts to ide</div>
</div>
</div>
<div class="clearfix"></div>
<hr class="mt-15 mb-15" />
</div>
</div>
</div>
</div>
</div>
</div>
<div class="modal fade" id="report" tabindex="-1" role="dialog" aria-hidden="true">
<div class="modal-dialog">
<div class="modal-content">
<form role="form" method="post" action="https://p.pdfkul.com/report/5a02405f1723ddd52c99261b" style="border: none;">
<div class="modal-header">
<button type="button" class="close" data-dismiss="modal" aria-hidden="true">×</button>
<h4 class="modal-title">Report arabic script language identifications using adaptive ...</h4>
</div>
<div class="modal-body">
<div class="form-group">
<label>Your name</label>
<input type="text" name="name" required="required" class="form-control" />
</div>
<div class="form-group">
<label>Email</label>
<input type="email" name="email" required="required" class="form-control" />
</div>
<div class="form-group">
<label>Reason</label>
<select name="reason" required="required" class="form-control">
<option value="">-Select Reason-</option>
<option value="pornographic" selected="selected">Pornographic</option>
<option value="defamatory">Defamatory</option>
<option value="illegal">Illegal/Unlawful</option>
<option value="spam">Spam</option>
<option value="others">Other Terms Of Service Violation</option>
<option value="copyright">File a copyright complaint</option>
</select>
</div>
<div class="form-group">
<label>Description</label>
<textarea name="description" required="required" rows="3" class="form-control"></textarea>
</div>
<div class="form-group">
<div style="display: inline-block;">
<div class="g-recaptcha" data-sitekey="6LeP2DsUAAAAAABvCByMZRCE253cahUVoC_jPUkq"></div>
</div>
</div>
<script src='https://www.google.com/recaptcha/api.js'></script>
</div>
<div class="modal-footer">
<button type="button" class="btn btn-default" data-dismiss="modal">Close</button>
<button type="submit" class="btn btn-primary">Save changes</button>
</div>
</form>
</div>
</div>
</div>
<!-- Modal -->
<div class="modal fade" id="login" tabindex="-1" role="dialog" aria-labelledby="myModalLabel">
<div class="modal-dialog" role="document">
<div class="modal-content">
<div class="modal-header">
<button type="button" class="close" data-dismiss="modal" aria-label="Close" on="tap:login.close"><span aria-hidden="true">×</span></button>
<h3 class="modal-title">Sign In</h3>
</div>
<div class="modal-body">
<form action="https://p.pdfkul.com/login" method="post">
<div class="form-group form-group-lg">
<label class="sr-only" for="email">Email</label>
<input class="form-input form-control" type="text" name="email" id="email" value="" placeholder="Email" />
</div>
<div class="form-group form-group-lg">
<label class="sr-only" for="password">Password</label>
<input class="form-input form-control" type="password" name="password" id="password" value="" placeholder="Password" />
</div>
<div class="form-group form-group-lg">
<div class="checkbox">
<label class="form-checkbox">
<input type="checkbox" name="remember" value="1" />
<i class="form-icon"></i> Remember Password </label>
<label class="pull-right"><a href="https://p.pdfkul.com/forgot">Forgot Password?</a></label>
</div>
</div>
<button class="btn btn-lg btn-primary btn-block" type="submit">Sign In</button>
</form>
</div>
</div>
</div>
</div>
<!-- Footer -->
<div class="footer-container" style="background: #fff;display: block;padding: 10px 0 20px 0;margin-top: 30px;">
<hr />
<div class="footer-container-inner">
<footer id="footer" class="container">
<div class="row">
<!-- Block footer -->
<section class="block col-md-4 col-xs-12 col-sm-3" id="block_various_links_footer">
<h4>Information</h4>
<ul class="toggle-footer" style="">
<li><a href="https://p.pdfkul.com/about">About Us</a></li>
<li><a href="https://p.pdfkul.com/privacy">Privacy Policy</a></li>
<li><a href="https://p.pdfkul.com/term">Terms and Service</a></li>
<li><a href="https://p.pdfkul.com/copyright">Copyright</a></li>
<li><a href="https://p.pdfkul.com/contact">Contact Us</a></li>
</ul>
</section>
<!-- /Block footer -->
<section id="social_block" class="col-md-4 col-xs-12 col-sm-3 block">
<h4>Follow us</h4>
<ul>
<li class="facebook">
<a target="_blank" href="" title="Facebook">
<i class="fa fa-facebook-square fa-2x"></i> <span>Facebook</span>
</a>
</li>
<li class="twitter">
<a target="_blank" href="" title="Twitter">
<i class="fa fa-twitter-square fa-2x"></i> <span>Twitter</span>
</a>
</li>
<li class="google-plus">
<a target="_blank" href="" title="Google Plus">
<i class="fa fa-plus-square fa-2x"></i> <span>Google Plus</span>
</a>
</li>
</ul>
</section>
<!-- Block Newsletter module-->
<div id="newsletter" class="col-md-4 col-xs-12 col-sm-3 block">
<h4>Newsletter</h4>
<div class="block_content">
<form action="https://p.pdfkul.com/newsletter" method="post">
<div class="form-group">
<input id="newsletter-input" type="text" name="email" size="18" placeholder="Entrer Email" />
<button type="submit" name="submit_newsletter" class="btn btn-default">
<i class="fa fa-location-arrow"></i>
</button>
<input type="hidden" name="action" value="0">
</div>
</form>
</div>
</div>
<!-- /Block Newsletter module-->
</div>
<div class="row">
<div class="bottom-footer">
<div class="container">
Copyright © 2024 P.PDFKUL.COM. All rights reserved.
</div>
</div>
</div>
</footer>
</div>
</div>
<!-- #footer -->
<script>
$(function () {
$("#document_search").autocomplete({
source: function (request, response) {
$.ajax({
url: "https://p.pdfkul.com/suggest",
dataType: "json",
data: {
term: request.term
},
success: function (data) {
response(data);
}
});
},
autoFill: true,
select: function (event, ui) {
$(this).val(ui.item.value);
$(this).parents("form").submit();
}
});
});
</script>
<!-- Google tag (gtag.js) -->
<script async src="https://www.googletagmanager.com/gtag/js?id=G-VPK2MQK127"></script>
<script>
window.dataLayer = window.dataLayer || [];
function gtag(){dataLayer.push(arguments);}
gtag('js', new Date());
gtag('config', 'G-VPK2MQK127');
</script>
</body>
</html>
<script data-cfasync="false" src="/cdn-cgi/scripts/5c5dd728/cloudflare-static/email-decode.min.js"></script>