A NEURAL MACHINE LANGUAGE TRANSLATION SYSTEM FROM GERMAN TO ENGLISH

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 04 | Apr 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 3094
A NEURAL MACHINE LANGUAGE TRANSLATION SYSTEM FROM GERMAN
TO ENGLISH
AKASH M1
1Department of Computer Science and Engineering, Anna University - CEG campus, Chennai India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Machine Translation is a Natural Language
Processing, attaining significant attention to fully automate
the system that can translate source content into the target
languages. The traditional systems focus on the word level
context while neglecting the sentence level context most
often, resulting in meaningless translation. This paper
focuses on the Transformer based MT which is used for
translation integrated with Fuzzy Semantic Representation
(FSR) for handling the occurrence of rare words in the
sentence level context. In the existing method there is a
mismatch of translation but the proposed system is more
superior due to the sentence context inclusion. The
algorithms proposed in this paper can be used for
multilingual translation systems. The FSR group rare words
together and represent sentence level context as Latent
Topic Representation (LTR) using convolutional neural
network. In particular, our algorithm can achieve a
significant performance for WMT En-De bilingual parallel
corpus and is used for translation handling the Out of
Vocabulary words using clustering of <UNK> tags. The
model performance is enhanced with hyper parameter
optimization obtaining a significantly high BLEU score, and
significantly low TER score with a be
Key Words: Natural Language Processing, Neural
Machine Translation(NMT), Transformer MT, Fuzzy
Semantic Representation(FSR), Latent Topic
Representation(LTR), Convolutional Neural Network
(CNN).
I. INTRODUCTION
Machine translation (MT) refers to fully automated
software which translates source language into target
language. Humans may use machine translation (MT) to
help them render speech and text into another language,
or the MT software may operate without human
intervention. Allowing a model to automatically (soft-
)search for parts of a source sentence that are relevant to
predicting a target word[1].
Automatic or machine translation is one of the most
challenging AI tasks given the fluidity of human language.
The classical approach of the rule-based machine
translation systems were used for translation, which were
replaced by statistical methods in the 1990s. In recent
times, deep neural network models achieve state-of-the-
art results for the language translation that is aptly named
as Neural Machine Translation (NMT).
Neural machine translation, or NMT in short, uses the
neural network models for machine translation. NMT
approaches are more powerful at capturing context
beyond phrase boundaries and were shown to better
exploit available training data[4]. The key benefit for
translation is that a single system can be trained directly
on source and target text, thereby not requiring the
pipeline of specialized systems which are used in
statistical machine translation(SMT). As such, NMT
systems are said to be end-to-end systems as only one
transformer-like model is required for the translation.
Transformer is an end-to-end model with encoder-
decoder architecture for seq-seq translation. The encoder
consists of a set of encoding layers that processes the
input iteratively one layer after another and the decoder
consists of a set of decoding layers that does the same
thing to the output of the encoder. Each of the encoder
layers function is to process its input to generate word
encodings (a.k.a word embeddings), containing
information about which parts of the input sentences are
relevant with each other. It passes its set of word
encodings to the next encoder layer as inputs to get a
better translation. Each decoder layer takes the output of
the encoder layer as input by taking all the word
encodings and processes them, using their incorporated
contextual information to generate an output
sequence[12].
To achieve this, each of the encoder and decoder layers
makes use of an attention mechanism (teacher enforcing),
which for each input word weighs the relevance of every
other input word and draws information (context) from
them accordingly to produce the output sentence. Each
layer of decoder also has an additional attention weight to
draw additional information from the outputs of previous
decoders, before the decoder layer draws information
from the encodings. Both the encoder and decoder layers
have a feed-forward neural network for additional
processing of the outputs, and contain residual
connections and layer normalization steps[13].
As in the current scenario there exist several languages in
the world. All the languages can’t be understood, so there
is a need for a language translator between the source and
target language. Natural Language Translation, which is a

NLP task, is the process of converting source language into
the target language using a Seq-Seq Encoder - Decoder
model[14]. A machine translator is a NLP task which
solves the problem of taking the input sentence of the
source language and giving it to the Neural Machine
Translator (NMT) for processing the language and giving
the output of the target language. Neural machine
translation (NMT) is an approach to machine translation
that uses an artificial neural network to predict the
likelihood of a sequence of words, typically modeling
entire sentences in a single integrated model. Topic
modeling techniques, such as Latent Dirichlet Allocation
(LDA) is very widely used for to capture semantics of the
sentence[2]. This is an effective approach for language
translation.
The proposed system uses a Fuzzy Semantic
Representation (FSR) of rare words to overcome the
<UNK> tag in the sentence along with the Latent Topic
Representation (LTR) using CNN to obtain the sentence
level context. FSR and LTR are integrated with the NMT
model to predict the target sentence given the source
sentence. Thus an efficient machine translation is
generated using the Transformer based encoder - decoder
Machine Translator[15].
II. RECENT WORKS
Kehai Chen et.al[3] have discussed sentence-level context
as a latent topic representation and designs a topic
attention to integrate source sentence-level topic context
information into both Attention based and Transformer
based NMT. LTR is represented using a Convolutional
Neural Network (CNN). The performance of NMT
improves by modeling source topics and translations
jointly. This study explored the dependence of source
topic information on its sentence-level context and
proposed a topic attention that integrated latent topic
representations into the existing NMT architecture to
improve translation prediction. To enhance target word
prediction it needs to exploit explicit source topic
information. It focuses on sentence-level context to learn
translation knowledge instead of document-level context
information, multi-source information, multi-model
information or syntax information.
Muyun Yang , Shujie Liu, Kehai Chen , et.al (2020)
proposes to build a fuzzy semantic representation (FSR)
method for rare words through a hierarchical clustering
method to group rare words together, and will be
integrated with the help of encoder–decoder framework.
Chinese-to-English sentence translation tasks confirm a
significant improvement in the translation quality for the
proposed method. To best preserve the semantic
information, this article proposes a hierarchical clustering
embedding approach to fuzzy semantic representation
(FSR) of rare words in NMT.
Xing Wang, Zhaopeng Tu, and Min Zhang (2015) have
explained a general framework for incorporating the SMT
word knowledge into NMT to alleviate above word-level
limitations. The NMT decoder makes more accurate word
prediction that is to be translated by referring to the SMT
word recommendations in both training and testing
phases. The SMT word predictions are used as prior
knowledge to adjust the NMT word generation probability,
which unitizes a neural network-based classifier to digest
the discrete word knowledge.
Rui Wang, Masao Utiyama, Andrew Finch and et.al. (2018)
have discussed the 4 goals for sentence level NMT domain
adaptation. First: The NMT’s internal sentence embedding
is exploited and the sentence embedding similarity is used
to select out-of-domain sentences which are close to the
in- domain corpus. Second: propose three sentence
weighting methods, i.e., sentence weighting, domain
weighting and batch weighting, to balance the data
distribution during NMT training. Third: In addition
propose dynamic training methods to adjust the sentence
selection and weighting during NMT training. Fourth: To
solve the multi-domain problem in a real-world NMT
scenario where the domain distributions of training and
testing data often mismatch, proposed a multi-domain
sentence weighting method to balance the domain
distributions of training data and match the domain
distributions of training and testing data.
Haipeng Sun , Rui Wang , Kehai Chen and et.al (2020) have
discussed about empirical relationships between
unsupervised neural machine translation (UNMT) and
unsupervised bilingual word embedding (UBWE) pre-
training and cross-lingual masked language model
(CMLM) pre-training. So a novel UNMT structure with
cross-lingual language representation agreement to
capture the interaction between UBWE / CMLM and
UNMT during UNMT training.
Kehai Chen , Tiejun Zhao, Muyun Yang and et.al (2018)
have given a novel neural approach to source dependence-
based context representation for translation prediction.
The proposed model is capable of not only encoding
source long-distance dependencies but also capturing
functional similarities to better predict translations. It is
incorporated into phrase-based translation (PBT) and
hierarchical phrase-based translation (HPBT) models. The
evaluation metrics used is Case Insensitive BLEU-4 metric
which outperforms all the baseline systems with better
translation. NMT still performs better than the proposed
model with a significantly greater BLEU score[5].
The methodologies and architectures employed in the
existing systems for machine translation were studied and
the issues in the existing systems include: inadequate
training of models, insufficient data, some models are built
for a specific language alone and are not applicable for

other common languages like English. Some models
translate sentences that resemble machine generated
sentences that do not sound like the language that humans
use to speak[6][7]. Machine Translation is needed to solve
this problem and the existing methods and related works
that were also studied. Transformer based NMT are more
natural and less artificial when compared to other models.
III. PROPOSED SYSTEM
The transformer is an end to end system having encoder-
decoder architecture. The encoder consists of a set of
encoding layers that processes the input sentence
iteratively one layer after another to produce sentence
vectors along with the rare words with the help of FSR
module and the decoder consists of a set of decoding
layers that does the same thing to the output sentence
vector of the encoder along with the topic vector to get a
more accurate translation.
The function of each encoder layer is to process its input
language sentence (English) to generate encodings in
vector form referred to as embeddings, containing
information about which parts of the inputs are relevant
to each other. It passes its set of encodings (obtained from
the previous encoding layer) to the next encoder layer as
inputs. Each decoder layer does the opposite of encoder,
taking all the encodings obtained from the encoder and
processes them, using their incorporated contextual
information to generate an output sequence. To achieve
this, each encoder and decoder layer makes use of an
attention mechanism (assigns weight to the input word
embeddings that is to be translated), which for each input,
weighs the relevance of every other input and draws
contextual information from them accordingly to produce
the output. Each decoder layer also has an additional
attention mechanism which draws information from the
outputs of previous decoders (assigns weight to the output
word embeddings that is being translated), before the
decoder layer draws information from the output of
encodings. Figure 1 shows the transformer integrating
LTR for sentence context along with LTR to handle rare
words which are OOV. Both the encoder and decoder
layers have a feed-forward neural network for additional
processing of the outputs, and contain residual
connections and layer normalization steps.
Figure 1: Transformer Architecture with FSR + LTR
The modules required to build an end-to-end Encoder
Decoder system for translation is described here. From
figure 1 the high level block diagram where the dataset of
bilingual parallel English-German corpus is preprocessed
to get a clean sentence by removing special characters and
punctuation marks. The cleaned dataset is splitted into
separate files of German sentences and English sentences.
The words which are OOV are given to the FSR most
commonly related together under a universal <UNK> tag.
The embedding module converts the sentence sequence
into its corresponding vector representation. The
positional encoding module uses sinusoidal functions to
indicate the position of the words in the sentence at both
source and target side. The context of the sentence is
handled as Latent Topic Representation which uses CNN
to extract the context of the sentence. The transformer
encoder and decoder which uses a self attention
mechanism to translate. Finally the decoded vector is
passed to the linear classifier and softmax functions.
Finally the google translator is integrated into our NMT
system to check the correctness of the predicted sentence
using the sentence prediction module.

A. TRANSFORMER ENCODER - DECODER
Each encoder consists of two major components: a self-
attention mechanism and a feed-forward neural network.
The self-attention mechanism takes in a set of input
encodings from the previous encoder and weighs their
relevance to each other to generate a set of output
encodings. The feed-forward neural network then further
processes each output encoding individually. These output
encodings are finally passed to the next encoder as its
input, as well as the decoders. The first encoder takes
positional information and embeddings of the input
sequence as its input, rather than encodings. The
positional information is necessary for the Transformer to
make use of the order of the sequence, because no other
part of the Transformer makes use of this[8].
Each decoder consists of three major components: a self-
attention mechanism, an attention mechanism over the
encodings, and a feed-forward neural network. The
decoder functions in a similar fashion to the encoder, but
an additional attention mechanism is inserted which
instead draws relevant information from the encodings
generated by the encoders. Like the first encoder, the first
decoder takes positional information and embeddings of
the output sequence as its input, rather than encodings.
Since the transformer should not use the current or future
output to predict an output though, the output sequence
must be partially masked to prevent this reverse
information flow. The last decoder is followed by a final
linear transformation and softmax layer, to produce the
output probabilities over the vocabulary[9].
Algorithm :
Continuous Bag of Words - Word2Vec model
Input: One hot vector of input/output sentence of size V
Output: Word Vector Representation - dependence of one
word on other words in sentence
Begin
1. Padding bits are added to the one hot vector to make
equi length sentence
2. Wvn is the weight matrix that maps input X to the
hidden layer (V*N dimensional matrix)
3. Wnv is the weight matrix that maps hidden layer output
to the final output layer (N*V dimensional matrix)
4. Hidden layer neurons copy the weighted sum of inputs
to the next layer
5. Non linearity Activation function of Softmax is used in
the Output layer
End
The encoder receives a source sentence x and encodes
each prefix using a recurrent neural network that
recursively combines embeddings xj for each word
position j:
where f is a non-linear function. Reverse encodings hj are
computed similarly to represent suffixes of the sentence.
These vector representations are stacked to form hj, a
representation of the whole sentence focused on position
j. The decoder predicts each target word yi sequentially
according to the distribution
where si is a hidden decoder state summarizing the prefix
of the translation generated so far, ci is a summary of the
entire input sequence, and g is another non-linear
function. Encoder and decoder parameters are jointly
optimized to maximize the log-likelihood of a training
corpus[5].
B. FSR - FUZZY SEMANTIC REPRESENTATION
Algorithm : Embedding Tree Construction
Begin
1. Train toolkit Word2Vec is used for Monolingual Data
2. Words OOV but appears >3 times are Clustered into M
classes by k-means
3. Class Embedding = Mean(Word Embeddings in the
class)
4. Each class has <UNKj> containing remaining words that
belong to this class (Cosine Similarity)
5. <UNKj> embedding = Mean(Remaining words in class)
End
The 3-level embedding tree is constructed based on the K-
means clustering and is categorised based on the class
which fits it into which is as shown in fig.2
Figure 2 : Classification of words in embedding tree
Algorithm : FSR Generation
Begin
1. Input Vector is Path from root to the rare word, to deal
with data sparseness
2. Differentiate class information and word information
for Rare words
3. Computes Source and Target Rare word Input vector

End
For the encoder, with previous encoder state hi−1, previous
input word embedding Exi−1, encoder computes the source
rare word input vector Exunk with attention mechanism
as
where Ep2 and Ep1 are the embeddings of word and class
node in the path, respectively, and Ux is a random
initialized weight matrix. Similar to the encoder, the target
rare word input vector Eyunk is based on previous
decoder state si−1, previous predicted word embedding
Eyi−1, previous context vector ci−1, and weight matrix Uy
The words which are out of vocabulary (OOV) are difficult
to translate and needs special attention for those words in
a sentence. The FSR module takes care of data sparseness
where the words which are
out of dictionary can be translated with better accuracy by
Differentiate class information and word information for
Rare words.
To integrate the proposed FSR into the existing
transformer-based NMT for improving target translation,
especially translations of rare words. Specifically, the
proposed FSR is only applied to the input representation
layer, whose output is the summation of word embeddings
and positional embeddings in the input sentence. Take the
source sentence as an example, if this word xj is not in-
vocabulary, the learned Exunk is used to replace the
original OOV vector;
if one word xj is in-vocabulary, its word vector is xj
The xj is, then, used to replace the existing xj to learn the
input representation.
C. LTR - LATENT TOPIC REPRESENTATION
Figure 3 shows all the layers of CNN model with (D X L)
dimensions of the sentence and extracts the meaning of
the sentence (context) for getting better translation
results.
Figure 3 : CNN model for LTR
The layers of CNN extracts the semantics of the sentence
based on the topics which are being defined initially. The
semantics closely matched with the most appropriate
topic.
Algorithm : Extracting Key Information
INPUT :
{T1, T2, T3,..., Tm} is M Topics
X={ x1, x2, x3 ,..., xj } is input sentence (Word embedded)
OUTPUT:
Topic distribution of a sentence - topic context vector
Begin
1. J*D input matrix layer is passed to 2M Convolution
Filters for 3 consecutive rows where D is the dimension of
the word vector and J is the sentence Length
2. The convolution Layer performs M row-wise max
operations to generate Max Pooling Layer
3. Max Pooling Layer undergoes Tanh function to generate
Output Layer
4. Finally T is mapped to {Key,Value} pair
End
Max-Pooling Layer: This takes row-wise maximum over
pairs of consecutive rows of L:
Pm = max(L2m−1, L2m), 1 ≤ m ≤ M
The resulting output feature map is P:
P = {P1, P2,..., PM}.
Note that a row pair-wise max-pooling operation is
applied to L2m−1 and L2m−1 to gain a D-dimensional topic

feature vector Pm. Compared with the max-pooling of the
sentence classification task, we use a D-dimensional topic
feature vector Pm to represent a topic in the input
sentence instead of a topic feature value. As we all know,
the vector representation has a better ability to encode
word or topic information. Meanwhile, the vector
representations of topics are also easier integrated into
the existing NMT architecture.
Output layer:This applies the tanh function over Pm to
obtain a LTR T m:
Tm = tanh(Pm),
T = {T1, T2,..., TM}.
The T is our proposed LTRs, which will later be used to
learn the topic context vector for NMT. The model
parameters for learning T are as the following:
η = {W1,W2,...,W2M; b1, b2,...,b2M}.
The CNN model layers extracts the key information as the
topic context vector for the NMT to work effectively.
D. DATASET DESCRIPTION
WMT '14 is open source data available on the Stanford
site. The modality of the dataset is text sentences. The
training data consists of 4.5 Million sentence pairs of both
German and English sentence pairs. The training and
validation is splitted in the ratio of 80% and 20%
respectively. The dataset is of rich text format and must be
preprocessed so that it can be fed into the model for
training. The dataset also includes the vocabularies of
words which occur more frequently, say top 50K frequent
words. The dictionary is formed by extracting from
alignment data[10][11].
IV. EVALUATION
The translated sentences were evaluated using Training
accuracy, BLEU and TER metrics. The loss and accuracy of
training is considered as one of the most important steps
towards evaluation. It is evident from Table 1 that
accuracy increases and loss decreases with the increase in
epochs. The loss is calculated based on the cross entropy.
Cross-entropy Loss= -∑ i=1 ∑ j=1 yi,j log(pi , j)
Accuracy = No. of correct predictions
Total no. of predictions
Loss = 1 - Accuracy
TABLE 1 : LOSS AND ACCURACY SCORES
Bilingual Evaluation Understudy (BLEU) is a score for
comparing target sentence translation with one or more
reference translations. The formula for calculating BLEU
Score is given as :
BLEU = BP . exp( ∑ Wn log pn )
Brevity Penalty (BP) = 1 if c > r
e(1-r/c) if c <= r
where r - count of words in a reference translation ;
c - count of words in a candidate translation.
Figure 4 shows that the training loss decreases as the
number of training epochs increases. At the range of 15 to
25 epochs the loss doesn't decrease significantly and
reaches the optimal point with a loss of almost 3% only.
Further increasing the epochs there is no further
improvement and the training time also increases with the
increase in the number of epochs.
Figure 4: Graph for Training Loss Score vs Epochs
Figure 5 shows that the training accuracy increases as the
number of training epochs increases. At the range of 15 to
25 epochs the accuracy doesn't increase significantly and
reaches the optimal point with an accuracy of almost 98%.
Further increasing the epochs there is no further
improvement and the training time also increases with the
increase in the number of epochs.

Figure 5: Graph for Training Accuracy Score vs Epochs
Figure 6 shows the comparative study of the loss and
accuracy for 5 different epochs, which shows the
improvements for epochs range between 15 to 25.
Figure 6: Bar Plot for Loss and Accuracy with Epochs
From Table 2, the BLEU score is calculated for the
translated sentence by our system with the google
translated sentence for the sample test cases and achieve a
maximum score of 0.84 for the third test case. We infer
that the short sentences get translated optimally by our
algorithm.
TABLE 2 : BLEU SCORE FOR 5 TEST CASES
TEST CASES BLEU SCORE
TEST CASE NO. BLEU SCORE
T1 0.67
T2 0.71
T3 0.84
T4 0.64
T5 0.8
From the figure 7, shows the BLUE evaluation metric for 5
different sample sentences from German to English
translation. The scores are obtained based on the above
mentioned BLUE score formula (Brevity Penalty) for the
translated sentences.
Figure 7: BLEU score for 5 sample translations
Translation Error Rate is the number of edits required to
change a system output into one of the references. The
formula for calculating TER Score is given as:
TER = Number of edits average
Number of reference words
TABLE 3 : TER SCORE FOR 5 TEST CASES
TEST CASES TER SCORE
TEST CASE NO. TER SCORE
T1 0.07
T2 0.08
T3 0.33
T4 0.04
T5 0.14
From table 3, the TER score is calculated for the translated
sentence by our system with the google translated
sentence for the sample test cases to minimize the
translation error rate of the system. Similarly the average
TER score for the entire corpus can be calculated.
From the figure 8, shows the TER evaluation metric for 5
different sample sentences from German to English
translation. From the graph above the Testcase T3 has
peak as the sentence length is long. So our model is limited
to short length translation sentences.

Figure 8: TER score for 5 sample translations
The performance of the system is evaluated based on the
BLEU score and the TER score. The aim is to maximize the
BLEU score so that the predicted translation which is
closely related to the reference translation (human
interference) and to minimize the TER score so that the
error in the translation becomes less. For sample 5 test
cases are evaluated using the above two scores and this
can be extended to the entire corpus if required.
For training, the total time taken by the batches per epoch
is tabulated below. Each epoch has 6 batches of size 100 to
be trained and the sum of all the time of batches per epoch
is calculated. The GPU with RAM of 25 GB is used to run
the model and the time taken per epoch is given in
seconds (sec). The training time for every 5 epochs
increases by a slight margin. From the training time for
50000 training sentences the time for every epoch
increases slightly and the average time taken for every
epoch is almost 45 seconds. To minimize the time per
epoch, a faster GPU with increased RAM size is needed.
The total time taken by the batches per epoch is tabulated
below. Each epoch has 6 batches of size 100 to be trained
and the sum of all the time of batches per epoch is
calculated and shown in Table 4. The GPU with RAM of 25
GB is used to run the model and the time taken per epoch
is given in seconds (sec). The training time for every 5
epochs increases by a slight margin.
TABLE 4 : MODEL TRAINING TIME VS EPOCHS
EPOCHS VS TIME
EPOCHS TIME (in sec)
1-5 189.2447
6-10 193.5873
11-15 193.3618
16-20 193.8892
21-25 194.2428
The plotting of Training Time for different epochs is
shown in fig. 9, for every 5 epochs the time is almost
constant but the time is largely dependent on the corpus
size. The time taken for training is directly proportional to
the corpus size and the number of epochs. The majority of
time is spent in model training for highly accurate
translation, so the optimal number of epochs must be
maintained for better efficiency.
Figure 9: Bar Plot of Training Time for different
epochs
V. RESULTS
Thus from the above tables we have the translated
sentences output by the model with the google translators
output. The end to end system for the translation of simple
german sentences to english sentences is obtained. Thus
the sample test cases with the Transformer based NMT
along with FSR to handle OOV words and LTR to handle
sentence context has better translation BLUE scores and
minimal TER scores in comparison with all other baseline
models. The preprocessing of sentences is done with the
help of regular expression by removing the special
characters from the sentence, for both german as well as
english sentences along with <start> and <end> tag to
denote the starting and ending of sentence. The words in a
sentence, which has been preprocessed are tagged with
numbers based on the ordering given by the dictionary,
for both german as well as english sentences. The
Transformer model along with FSR and LTR is trained
with 25 epochs to get optimal results for the translation.
The training along with the accuracy and loss for each
batch in epochs are done. The WordtoVec model
(pretrained model) is used to obtain the word embeddings
(encoded vector). The German to English translation for
the trained data is tested and compared with the google’s
translation for validating the translated output. The
attention graph is plotted which is similar to heatmap (i.e)
the brightest color is taken as the matched translated
word. The transformer model uses the self attention
mechanism for the translation.

VI. CONCLUSION
In this work, building an end to end Language Translation
system which uses a Transformer based NMT to convert
German Language sentences to English Language
sentences. The context of a sentence is taken into account
so that the target sentence translation will be of the same
meaning of the source sentence. The Self Attention
mechanism used in the transformer predicts a better
translated sentence. The rare words which are OOV in the
sentence are handled using FSR with the help of
hierarchical clustering which is integrated with the
transformer model. The context of the sentence acts as the
LTR which uses CNN is also integrated within the system.
The Sequence to Sequence system for the dataset takes a
significant time for modeling the system. The experiments
indicated that the proposed NMT system achieves good
performance with a better BLEU score compared to
conventional MT systems and minimizes the TER
significantly.
VII. FUTURE WORKS
The future work for this project lies in the module of
machine translator, where more attention is needed to
generate more accurate translation as in accordance with
the human translated output rather than google translated
output. This work will be related to multiple reference
translation using the GAN model so that it achieves a close
match with human translation. These translation systems
can be applied to many other applications like chatBot
systems, Helpdesk applications etc.
Thus we have translated sentences from German to
English with the help of Google Translation for the limited
sentence sizes. In the near future we are planning to
extend this system for lang sentences along with
multilingual translations from any to any known scripted
languages with the help of google translation package.
REFERENCES
[1] Bahdanau, Dzmitry, Kyunghyun Cho, and Yoshua
Bengio. "Neural machine translation by jointly
learning to align and translate." arXiv preprint
arXiv:1409.0473 (2014).
[2] Cao, Ziqiang, et al. "A novel neural topic model and its
supervised extension." Twenty-Ninth AAAI
Conference on Artificial Intelligence. 2015..
[3] Chen, Kehai, et al. "A neural approach to source
dependence based context model for statistical
machine translation." IEEE/ACM Transactions on
Audio, Speech, and Language Processing 26.2 (2017):
266-280.
[4] Chen W., Matusov E., Khadivi S., and Peter J., “Guided
alignment training for topic-aware neural machine
translation,” in The 38th Annual Meeting and
Symposium of the Antenna Measurement Techniques
Association, vol. 1, Austin, Texas, November 2016, pp.
121–134.
[5] Chitnis R. and DeNero J., “Variable-length word
encodings for neural translation models,” in
Proceedings of the 2015 Conference on Empirical
Methods in Natural Language Processing, 2015, pp.
2088–2093.
[6] Chung J., Cho K., and Bengio Y., “A character-level
decoder without explicit segmentation for neural
machine translation,” in Proceedings of the 54th
Annual Meeting on Association for Computational
Linguistics. Association for Computational Linguistics,
2016, pp. 1693–1703.
[7] Fadaee M., Bisazza A., and Monz C., “Learning topic-
sensitive word representations,” in Proceedings of the
55th Annual Meeting of the Association for
Computational Linguistics (Volume 2: Short Papers).
Vancouver, Canada: Association for Computational
Linguistics, Jul. 2017, pp. 441–447.
[8] Haipeng Sun,Rui Wang, Kehai Chen, Masao Utiyama,
Eiichiro Sumita, and Tiejun Zhao."Unsupervised
Neural Machine Translation With Cross-Lingual
Language Representation Agreement." IEEE/ACM
Trans on audio,speech,and language processing,Vol.
28, 2020.
[9] Jean S., Cho K., Memisevic R.,and Bengio Y.,“On Using
Very Large Target vocabulary for neural machine
translation,” in Proc. 53rd Annu. Meet. Assoc. Comput.
Linguistics 7th Int. Joint Conf. Natural Lang. Process.,
Jul. 2015, pp. 1–10.
[10] Kehai Chen, Rui Wang, Masao Utiyama, Eiichiro
Sumita, and Tiejun Zhao."Neural Machine Translation
with Sentence-level Topic Context."IEEE/ACM Trans
on audio,speech,and language processing,
doi:10.1109/TASLP.2019.2937190.
[11] Kim Y., “Convolutional neural networks for sentence
classification,” in Proceedings of the 2014 Conference
on Empirical Methods in Natural Language Processing
(EMNLP). Doha, Qatar: Association for Computational
Linguistics, Oct. 2014, pp. 1746–1751.
[12] Muyun Yang , Shujie Liu, Kehai Chen ,Hongyang Zhang,
Enbo Zhao, and Tiejun Zhao."A Hierarchical Clustering
Approach to Fuzzy Semantic Representation of Rare
Words in Neural Machine Translation." IEEE Trans.on
fuzzy logic, Vol. 28, no. 5, May 2020.

[13] Rui Wang, Masao Utiyama, Andrew Finch, Lemao Liu,
Kehai Chen, and Eiichiro Sumita."Sentence Selection
and Weighting for Neural Machine Translation
Domain Adaptation."IEEE/ACM Trans. on audio,
speech, and language processing, doi:
10.1109/TASLP.2018.2837223.
[14] Snover, Matthew, et al. "A study of translation edit rate
with targeted human annotation." Proceedings of
association for machine translation in the Americas.
Vol. 200. No. 6. 2006.
[15] Xing Wang, Zhaopeng Tu, and Min Zhang"
Incorporating Statistical Machine Translation Word
Knowledge into Neural Machine Translation".Journal
of latex class files, Vol. 14, no. 8, August 2015.

A NEURAL MACHINE LANGUAGE TRANSLATION SYSTEM FROM GERMAN TO ENGLISH

More Related Content

Similar to A NEURAL MACHINE LANGUAGE TRANSLATION SYSTEM FROM GERMAN TO ENGLISH (20)

More from IRJET Journal (20)

Recently uploaded (20)

A NEURAL MACHINE LANGUAGE TRANSLATION SYSTEM FROM GERMAN TO ENGLISH