See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/362581063

Implementation of NLP based automatic text summarization using spacy

Article in International Journal of Health Sciences · August 2022

DOI: 10.53730/ijhs.v6nS5.10574

CITATIONS READS

9 1,411

6 authors, including:

Achyutha Prasad N Piyush Kumar Pareek

East West Institute of Technology Nitte Meenakshi Institute of Technology
49 PUBLICATIONS 581 CITATIONS 63 PUBLICATIONS 709 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Achyutha Prasad N on 22 September 2022.

The user has requested enhancement of the downloaded file.

How to Cite:
Prakash, N. C. P., Narasimhaiah, A. P., Nagaraj, J. B., Pareek, P. K., Maruthikumar, N. B., &
Manjunath, R. I. (2022). Implementation of NLP based automatic text summarization using
spacy. International Journal of Health Sciences, 6(S5), 7508–7521.
https://doi.org/10.53730/ijhs.v6nS5.10574

Implementation of NLP based automatic text

summarization using spacy

Nayana Cholanayakanahalli Prakash

Post Graduation Student, Master of Technology, Department of Computer
Science and Engineering, East West Institute of Technology, Bengaluru,
Karnataka, India; Pincode: 560091
Corresponding author email: nayanaewit@gmail.com

Achyutha Prasad Narasimhaiah

Professor, Department of Computer Science and Engineering, East West Institute
of Technology, Bengaluru, Karnataka, India; Pincode: 560091
Email: achyuth001@gmail.com

Jagadeesh Bettakote Nagaraj

Assistant Professor, Department of Computer Science and Engineering, East West
Institute of Technology, Bengaluru, Karnataka, India; Pincode: 560091
Email: jagadeeshbn001@gmail.com

Piyush Kumar Pareek

Professor, Department of Computer Science and Engineering & Head of IPR Cell,
Nitte Meenakshi Institute of Technology, Bengaluru, Karnataka, India; Pincode:
560064
Email: piyush.kumar@nmit.ac.in

Nalini Bpalya Maruthikumar

Assistant Professor, Department of Computer Science and Engineering, East
West Institute of Technology, Bengaluru, Karnataka, India; Pincode: 560091
Email: nalinibm03@gmail.com

Ramya Iyaravally Manjunath

Assistant Professor, Department of Computer Science and Engineering, East
West Institute of Technology, Bengaluru, Karnataka, India; Pincode: 560091
Email: ramyaim31@gmail.com

Abstract---The amount of data on the Internet has increased

exponentially over the past decade. Therefore, we need a solution that
converts this massive amount of raw information into useful
information that the human brain can understand. One such common
technique in research that helps when dealing with large amounts of
data is text summarization. Automatic summarization is a well-known

International Journal of Health Sciences ISSN 2550-6978 E-ISSN 2550-696X © 2022.

Manuscript submitted: 9 April 2022, Manuscript revised: 18 June 2022, Accepted for publication: 27 July 2022
1
 7509

approach to reduce documents to key ideas. This works by storing

important information by creating a shortened version of the text. Text
summaries are divided into extraction and abstraction methods. The
extraction summary method minimizes the summarization burden by
selecting a subset of relevant sentences from the actual text. There are
many methods, but researchers specializing in natural language
processing (NLP) are particularly attracted to the extraction method.
The meaning of the sentence is calculated using linguistic and
statistical features. In this work, extractive and abstract methods for
summarizing texts were examined. This white paper uses a spacey
algorithm to analyze the above methods, resulting in fewer iterations
and a more focused summary.

Keywords---Empirical Methods, Text Summarization, Extraction,

Abstraction, Reinforcement Learning, Supervised, Unsupervised, NLP,
Spacy Algorithms.

Introduction

Summarizing large texts remains an open topic in natural language processing.

Automatic text summarization is used to summarize large documents. Text
summary is the process of using software to simplify a text document and create
an overview of the gist of the original document. Text summarization is a
technique that software uses to shrink a text document to create a summary or
synopsis of the original document. Summary is performed to highlight important
parts of the text. Text summaries can be categorized based on the input type.
Single low-input document in a text context. In such cases, a simple summary
model is created. Several documents, the input of which can be relatively long.
The more text you have, the more semantic links are generated, which adds to the
complexity here. Summarizers can be categorized as general based on their
purpose. It is abstract and the model forms its own phrases and sentences to
provide a more consistent summary. In this case, the model processes the input
without bias or prior knowledge. Domain specific. The model uses domain
information to create a more accurate summary based on known facts. Query
based. The summary contains only known answers to natural language questions
about the input text. Based on output type, totalizers can be categorized as
follows: Extract. Select a key phrase from the input text to create a summary that
humans generate. Creating an abstract summary is generally a more complex
task than the extraction method. Therefore, apart from recent advances in the use
of neural networks, driven by advances in neural machine translation and
sequence models, they are still far from reaching the human level. Text
summarization applications include media monitoring, search engine marketing,
internal document workflows, financial research, social media marketing, and
assisting people with disabilities.

Method

This segment reviews a number of excellent text summaries as shown in Figure 1.

7510

Unsupervised Extracts

Unsupervised summarization techniques mean creating summaries from a given

document without using previously identified groups or classifications. There are
three ways to do this: first graph-based, latent variables, and finally term
frequency. These are easy to implement and give satisfactory results. Some of the
surveys conducted are listed below. Hernándecetal. presented a solution for
selecting sentences in an extract summary using KMeans clustering. This is a big
disadvantage. The first step is to remove stop words, hyphens and extra spaces.
This is called input text preprocessing.

The next step is to use n-grams to select features and use boolean weight (BOOL),
term frequency (TF), inverse document frequency (IDF), or TFIDF to find weights.
The next step is to apply KMeans to the clustering set. KMeans is an iterative
process that plots values up to the nearest centroid (the average of all values) and
then calculates a new centroid. In the proposed method, the first sentence is
considered as the baseline and the similarity between the sentences is plotted
using Euclidean distance. After clustering is done using the K-cluster, the set that
is closest to the centroid (aka the most representative set) is selected. The
proposed method achieves better results than other prior art methods.

Figure 1: Classification of text summaries

Joshi et al. proposed an unsupervised framework for extracting text

summarization of a single document called SummCoder. After preprocessing,
SummCoder uses a jump thinking model to transform the sentence into a fixed-
length vector. To create a summary, sentences are selected considering three
scores: Sentence Content Relevance Metric (scoreContR), Sentence Novelty Metric
(scoreNov), and Sentence Position Relevance Metric (scorePosR). After all points
have been calculated, the final score and relative score will be calculated. Finally,
summaries can be generated by first sorting by descending relative rank and then
by appearance in the input text.

El-Kassas et al. announced a single document and graph-based extraction system

called EdgeSumm. In the proposed method, a pretreatment and a lemming are
 7511

carried out first. Then a graph of the textual representation is created with nouns
as nodes and non-noun words as edges. There are "S#" and "E#" nodes that
indicate the beginning and end of the sentence. For each node, the weight is
calculated by counting the number of times it occurs. When choosing a sentence,
it is assumed that all nouns represent different topics. First, search for the most
common words and phrases and make a list of selected nodes and edges. To
select the source and destination nodes, the score must be higher than the
average score for all nodes. To select an edge, you must select both the source
node and the target node. If the candidate summary (the summary generated by
the algorithm) exceeds the user limit, the candidate summary will be scored and
ranked in ascending order. Then apply KMeans clustering to group similar
statements and select the top statements from each cluster to create the final
summary.

Zheng & Lapata, proposed a position-enhanced centrality-based summary

(PacSumm). Uses a graph-based rank algorithm. A set is a node and an edge
shows the relationship between the nodes. A bi-directional encoder representation
of Transformers (BERT) was used to map the set. The BERT pre-training has two
tasks. The first task is masked language modeling, where the sentences are
tokenized with the left and right sentences. The second task is sentence
identification, which identifies the relationship between two sentences. To fine-
tune the BERT, five negative samples are reported for each positive sample. After
finding the representation of all sets, use the pairwise dot product to create an
unnormalized matrix. Sets are selected based on this matrix.

Vanetiketal proposed a weighted compression model to extract important

information from text. In the proposed model, this is done by reducing the
sentence by repeatedly deleting the elementary discourse unit (EDU). First, each
word is given a non-negative weight. Weights are assigned using the Gillic and
Favre and McDonald extraction models. The next step is to select and delete the
EDU. The list of EDUs is created using the constituency-based syntax tree.
Excluded from the list are EDUs, which may make an ungrammatical statement
when deleted. Everything else is removed and the "important" EDU weights are
calculated and sorted. During summary creation, EDUs are selected based on
maximum weight-to-cost ratio and cannot exceed the length of the summary.

Ozsoy & Alpaslan introduced Latent Semantics (LSA) for text summarization. This
is an algebraic static method for finding hidden logical patterns between words
and sentences. The input matrix is created to display the text. Rows represent
words and columns represent sentences. The cell shows the TF-IDF value of the
word. Single Value Decomposition (SVD) is used to model the relationship
between words and sentences. The result of the SVD is useful for selecting
instructions using the cross method. The set with the longest vector is selected.
7512

Discussion

A. NLP and knowledge base

The study of natural language processing (NLP) dates from the 1950s. NLP
understand and processes languages using sentence grammar, ontology,
language models, analysis trees and similar methods. NLG (Natural Language
Generation) does the opposite and generates natural language from machine
rendering. NLP / NLG-based summaries are sometimes referred to as "semantics-
based or ontology-based" (Allahyari et al., 2017) rather than "knowledge-based.
Before the advent of deep learning systems, NLP and ontology-based solutions
have been the most common ways to do abstract transformations. For example,
sentences can be grouped by mechanical conjunction rules. This type of
abstraction is primarily grammatical and may not provide integration of
document ideas. Think of this "abstract light".

Some researchers have combined NLP with deep learning to "encode" "linguistic
information" such as Part-of-Speech (POS) tags and Named Entity Extraction
(NER) tags as a lexical function as part of an encoder-decoder neural network. .
(Zhou, Yang, Wei, Tan & Bao, 2017). I agree with Allahyari and others. (2017)
"The step to building a more accurate summarization system is to combine the
summarization method with knowledge-based and semantics-based ontology-
based summarization." The trend that can be seen in the comparison matrix is
away from NLP and towards deep learning

B. Spacy

SpaCy is a free open-source natural language processing library in Python

programming languages. Used primarily in production software development,
spaCy also supports deep learning workflows via PyTorch and TensorFlow
statistical models. SpaCy provides fast and accurate parsing, named entity
recognition, and easy access to word vectors. You can use the default word vector
or replace it with another word vector. SpaCy also features tokenization, sentence
boundary detection, part-of-speech tagging, parsing, built-in word vectors, and
very accurate placement on the original string.

Fig 2: Text Summarization

Text summarization procedure:

1. Text cleaning: Remove stop words, punctuation, and lowercase letters in

words.
2. Working tokenization: Tokenize all words from a sentence.
 7513

3. Table of Word Frequency each word Frequency is calculated and divide

maximum frequency with each frequency to get the normalized word
frequency count.
4. Sentence tokenization: As per Frequency of the sentence summary will done.

The other spacy application we are investigating consists of automatic

summarization sentence extraction. In a way, you can think of the sentence
extraction problem as similar to keyword extraction. Both applications aim to
identify sequences that are more "representative" of a given text. In keyword
extraction, candidate text units consist of words, while sentence extraction
processes the entire sentence. Text Rank proves suitable for this type of
application as it accepts a recursively computed per-text ranking based on
information gleaned from the entire text.

The evaluation according to the essential approach is often carried out by

comparing the results of the machine overview with the expert's ideal overview.
Substantial assessments are made by measuring grammar, verbosity, and
consistency. This measurement is very rare because it is a very rare study that
poses grammar and consistency issues. Find out how many ideal sentences there
are in the automated machine summary and evaluate them from a sentence
extraction perspective. The evaluation was carried out using the measurement
methods Precision, Recall and f-Score / f-Measure. To evaluate the content,
compare the actual words of the sentence, not the entire sentence. The evaluation
was made by measuring the similarity of rouge, pyramid and cosine. The benefit
of scoring is that human extracts can be compared to an automatic
summarization engine

Use a human summary with new sentences and paraphrases. Another

assessment is a task-based approach that measures the performance of the
automatic summary engine by using the summary for specific tasks such as
answering questions and document categories.
7514

Results and Discussions

Architecture of the Project:

Fig 3: Architecture of the Project

The project architecture is shown in the figure 3 above. As you can see, the text
document is uploaded to the application first. The text document is then
preprocessed, including the removal of stop words and punctuation, by finding
the word Frequency and the Sentence Frequency. Finally, create a text summary.

Document pre-processing

Due to the Excess sources of information in today's world, the input documents
we receive may not be in the correct English format, which may contain audio.
Sounds include various special characters, unwanted spaces, newlines, stops,
and more. Therefore, perform the following tasks on the input file to get only the
useful parts of the document.
Step 1: All line breaks are removed.
Step 2: All corner brackets and special numbers are removed.
Step 3: All commas, extra spaces and repeating sentences are removed.

Removal of stop words

In this step it will remove all subtitles from your input according to your native
language. These stop words do not provide reliable information about a particular
context. It does not convey any information about this emotion as it builds a
collection of emotions like "is", "am", "who" to create an illustration.

Tokenization

Previously, sentences were split into several words. Basically, this token model is
used to do the activity in the form of a pipelined NLP natural language processing
process. This is useful at two stages, word level and sentence level. The first is a
standard word mark that restores a set of words in a given sentence.
 7515

Extraction of important sentences

We need a way to verify the value of the text in the scroll. The following
calculations are performed to extract the key phrase from the document and the
same is shown in figure 4 & 5.
Step 1: Frequency of all words in the previously reviewed text is calculated.
Step 2: The weight of each word by dividing the frequency of the words by the
maximum frequency is calculated.
Step 3: Review all key phrases for the specified input.
Step 4: The sentence's score by adding the weighted frequencies of the words
contained in the sentence is calculated.
Step 5: Sort the set's token list in descending order based on points
Step 6: Remove the "n" statement from the token list

Figure 4: Preprocessing

Fig 5: Viewing the Summary

7516

Conclusion

Automatic text summarization is an interesting academic topic with a wide range

of commercial applications. By reducing huge amounts of information into short
bursts, summaries are useful in a variety of downstream applications such as
news summaries, reporting, news summaries and headlining. There are two types
of aggregation algorithms that are most commonly used. The extraction summary
method starts with rearranging and copying the passages from the source
material. Second, the abstract summary approach creates new phrases by
rephrasing or inserting terms not found in the original text. Most of the research
to date has been extractive due to the difficulty of abstract summarization. The
extraction approach is more convenient as it guarantees grammar and accuracy
by copying a large chunk of text from the source document. On the other hand,
advanced skills such as paraphrasing, generalizing, and assimilation of real-world
knowledge are only possible with abstract frameworks and are required for high-
quality summarization. Despite the fact that abstract summarization is a more
difficult task, there has been some success thanks to recent advances in deep
learning.

Acknowledgments

I extend my deep sense of sincere gratitude to Dr. Channakesavalu K, Principal,

East West Institute of Technology, Bengaluru, for having permitted to carry out
the survey on “NLP Based Automatic Text Summarization using Spacy”
successfully.
I express my heartfelt sincere gratitude to my guide Dr. Achyutha Prasad N,
Head, Department of Computer Science and Engineering, East West Institute of
Technology, Bengaluru for his valuable guidance, encouragement and
suggestions.

I would like to express my sincere thanks to my internal co-guide Jagadeesh B N,

Assistant Professor, Department of Computer Science and Engineering, East West
Institute of Technology, Bengaluru for his valuable guidance, encouragement and
suggestions.

I would like to thank all the Teaching, Technical faculty and supporting staff
members of Department of Computer Science and Engineering, East West
Institute of Technology, Bengaluru, for their valuable suggestions and support.

Finally, I would like to thank my Parents for their support.

References

A. P. N and C. D. Guruprakash, "A Relay Node Scheme for Energy Redeemable

and Network Lifespan Enhancement," 2018 4th International Conference on
Applied and Theoretical Computing and Communication Technology (iCATccT),
Mangalore, India, 2018, pp. 266-274.
Abbasi-ghalehtaki, R., Khotanlou, H., and Esmaeilpour, M. (2016). Fuzzy
evolutionary cellular learning automata model for text summarization. Swarm
and Evolutionary Computation, 30:11–26.
 7517

Abdi, A., Shamsuddin, S. M., and Aliguliyev, R. M. (2018). Qmos: Query-based

multi-documents opinion-oriented summarization. Information Processing &
Management, 54(2):318–338. J. Clerk Maxwell, A Treatise on Electricity and
Magnetism, 3rd ed., vol. 2. Oxford: Clarendon, 1892, pp.68–73.
Achyutha Prasad, N., Guruprakash, C.D., 2019. A relay mote wheeze for energy
saving and network longevity enhancement in WSN. International Journal of
Recent Technology and Engineering 8, 8220–8227. doi:10.35940/ijrte.C6707.
Achyutha Prasad, N., Guruprakash, C.D., 2019. A relay node scheme of energy
redeemable and network lifespan enhancement for wireless sensor networks
and its analysis with standard channel models. International Journal of
Innovative Technology and Exploring Engineering 8, 605–612.
Achyutha Prasad, N., Guruprakash, C.D., 2019. A two hop relay battery aware
mote scheme for energy redeemable and network lifespan improvement in
WSN. International Journal of Engineering and Advanced Technology 9, 4785–
4791. doi:10.35940/ijeat.A2204.109119.
Achyutha, P. N., Hebbale, S., & Vani, V. (2022). Real time COVID-19 facemask
detection using deep learning. International Journal of Health Sciences, 6(S4),
1446–1462. https://doi.org/10.53730/ijhs.v6nS4.6231.
Automatic text summarization using reinforcement learning with embedding
features. In Proceedings of the Eighth International Joint Conference on
Natural Language Processing (Volume 2: Short Papers), volume 2, pages 193–
197.
Chetana Srinivas, Ambrish G, Bharathi Ganesh, Anitha Ganesh, Dhanraj, Kiran
M, “Logistic Regression Technique for Prediction of Cardiovascular Disease”,
International Conference on Intelligent Engineering Approach,(ICIEA) India,
12th February 2022.
Chetana Srinivas, Ambrish G, Supritha N, Bharathi G, Anitha G, “Survey on
Recent Trends in Machine Learning and Deep Learning in Healthcare”,
International Conference on Recent Trends in Machine Learning and
Computing System,(RTMCS) India, 17th -18th December 2021.
Chetana Srinivas, Nandini Prasad K S,"A Comparative study on Medical Image
Processing Using Big Data Analytics Frameworks”, 2018 Third International
Conference on Electrical, Electronics, Communication, Computer Technologies
and Optimization Techniques (ICEECCOT), Mysuru, India, 2018.
Chetana Srinivas, Nandini Prasad K S,” A Comparative Study on Different Types
of Image Pre-processing Methods for Noise Removal”, Internal Journal of
Computing, Communication & Networking (IJCCN), ISBN: 2319-2720, Vol.7,
Issue 2, April 2018.
Chetana Srinivas, Nandini Prasad K. S., Mohammed Zakariah, Yousef Ajmi
Alothaibi , Kamran Shaukat , B. Partibane, and Halifa Awal, “Deep Transfer
Learning Approaches in Performance Analysis of Brain Tumor Classification
Using MRI Images”, Hindawi Journal of Healthcare Engineering Volume 2022,
Article ID 3264367, 17 pages https://doi.org/10.1155/2022/3264367.
Daum´e III, H. and Marcu, D. (2004). A tree-position kernel for document
compression. In Proceedings of DUC2004.
Dr.Balakrishna R, Piyush Kumar Pareek et al, ’Data Mining for Healthy Tomorrow
with the implementation of Software Project Management technique’, Springer
AISC Series/ SCOUPS INDEXED JOURNAL, Paper Id : IT -187-ICPCIT2015,
June 2015.
7518

Dr.Balakrishna R, Piyush Kumar Pareek et al, ’Study on Six Sigma approach to

improve the quality of process outputs in business processes in Small &
Medium Level Software Firms’ Springer AISC Series/ SCOUPS INDEXED
JOURNAL, Paper Id : IT -221-ICPCIT2015.
Dr.Piyush Kumar Pareek et al, ‘A survey on approaches for predicting
performance of students’,International Journal of Engineering Research and
Science, ISSN No.2395-6992 Paper Id:IJOER-Jun-2016-25.
Dr.Piyush Kumar Pareek et al, ‘A survey on Long term product planning and
requirements prioritization to customer value creation’, International Journal
of Engineering Research and Science, ISSN No.2395-6992 Paper Id: IJOER-
Jun-2016-27.
Dr.Piyush Kumar Pareek et al, ‘Education Data Mining –Perspectives of
Engineering Students ’, International Journal of Innovative Research in
Computer Science & Technology (IJIRCST), ISSN: 2347-5552, Volume-4, Issue-
5, September-2016.
Edmundson, H. P. (1969). New methods in automatic extracting. Journal of the
ACM, 16(2):264–285.
Hebbale, S., Marndi, A., Achyutha, P. N., Manjula, G., Mohan, B. R., &
Jagadeesh, B. N. (2022). Automated medical image classification using deep
learning. International Journal of Health Sciences, 6(S5), 1650–1667.
https://doi.org/10.53730/ijhs.v6nS5.9153.
Hebbale, S., Marndi, A., Manjunatha Kumar, B. H., Mohan, B. R. ., Achyutha, P.
N., & Pareek, P. K. (2022). A survey on automated medical image
classification using deep learning. International Journal of Health Sciences,
6(S1), 7850–7865. https://doi.org/10.53730/ijhs.v6nS1.6791.
Hovy, E. and Lin, C. Y. (1999). Automated text summarization in summarist. In
Mani, I. and Maybury, M. T., editors, Advances in Automatic Text
Summarization, pages 81–94. MIT Press.
Jipeng, T., Neelagar, M. B., & Rekha, V. S. (2021). Design of an embedded control
scheme for control of remote appliances. Journal of Advanced Research in
Instrumentation and Control Engineering, 7(3 & 4), 5-8.
Kadakadiyavar, S., Prasad, A. N., Pareek, P. K., Vani, V., Rekha, V. S., &
Nirmala, G. (2022). Recognition efficiency enhancement of control chart pattern
using ensemble MLP neural network. International Journal of Health Sciences,
6(S3), 4295–4306. https://doi.org/10.53730/ijhs.v6nS3.6851.
Kalshetty, J. N., Achyutha Prasad, N., Mirani, D., Kumar, H., & Dhingra, H.
(2022). Heart health prediction using web application. International Journal of
Health Sciences, 6(S2), 5571–5578.
https://doi.org/10.53730/ijhs.v6nS2.6479.
Knight, K. and Marcu, D. (2000). Statistics-based summarization - step one:
Sentence compression. In AAAI/IAAI, pages 703–710.
Kogilavani, A. and Balasubramanie, P. (2010). Clustering based optimal summary
generation using genetic algorithm. In Communication and Computational
Intelligence (INCOCCI), 2010 International Conference on, pages 324–329.
IEEE. Lee, G. H. and Lee, K. J. (2017).
Kupiec, J., Pedersen, J., and Chen, F. (1995). A trainable document summarizer.
In Proceedings SIGIR ’95, pages 68–73, New York, NY, USA.
Lebanon, G. (2006). Sequential document representations and simplicial curves.
In Proceedings of the 22nd Conference on Uncertainty in Artificial Intelligence.
 7519

Luhn, H. P. (1958). The automatic creation of literature abstracts. IBM Journal of

Research Development, 2(2):159–165.
Mani, I. and Bloedorn, E. (1997). Multi-document summarization by graph search
and matching. In AAAI/IAAI, pages 622–628.
Manjunatha Kumar, B. H., Achyutha , P. N., Kalashetty, J. N., Rekha, V. S.,
& Nirmala, G. (2022). Business analysis and modelling of flight delays
using artificial intelligence. International Journal of Health Sciences, 6(S1),
7897–7908. https://doi.org/10.53730/ijhs.v6nS1.6735.
Mehdi Allahyari and Krys Kochut. 2015. Automatic topic labeling using ontology-
based topic models. In Machine Learning and Applications (ICMLA), 2015 IEEE
14th International Conference on. IEEE, 259–264.
Mr. Piyush Kumar Pareek, Dr. A. N. Nandakumar, Lean software development
Survey on Agile and Lean usage in small and medium level firms in Bangalore,
International Journal of Advanced Research in Computer Science and Software
Engineering , Volume 4, Issue 12, December 2014 , ISSN: 2277 128X .pp 1-7
Impact Factor : 2.08.
Mr.Piyush Kumar Pareek, Dr. A. N. Nandakumar, ’Lean software development
Survey on Benefits and challenges in Agile and Lean usage in small and
medium level firms in Bangalore’ , International Journal of Advanced Research
in Computer Science and Software Engineering , Volume 4, Issue 12,
December 2014 , ISSN: 2277 128X .pp 1-11.
N. A. Prasad and C. D. Guruprakash, "An ephemeral investigation on energy
proficiency mechanisms in WSN," 2017 3rd International Conference on
Applied and Theoretical Computing and Communication Technology (iCATccT),
Tumkur, 2017, pp. 180-185.
N. G and G. C. D, "Unsupervised Machine Learning Based Group Head Selection
and Data Collection Technique," 2022 6th International Conference on
Computing Methodologies and Communication (ICCMC), 2022, pp. 1183-1190,
doi: 10.1109/ICCMC53470.2022.9753995.
Narayan, S., Cohen, S. B., and Lapata, M. (2018). Ranking sentences for
extractive summarization with reinforcement learning. arXiv preprint
arXiv:1802.08636.
Oufaida, H., Nouali, O., and Blache, P. (2014). Minimum redundancy and
maximum relevance for single and multidocument arabic text summarization.
Journal of King Saud University-Computer and Information Sciences,
26(4):450– 461.
Parveen, D., Mesgar, M., and Strube, M. (2016). Generating coherent summaries
of scientific articles using coherence patterns. In Proceedings of the 2016
Conference on Empirical Methods in Natural Language Processing, pages 772–
783.
Piyush Kumar Pareek & Dr. A. N. Nandakumar, ’To Implement Lean software
development frame- work for minimizing waste in terms of non-value added
activities’, Research Publishing, Jain University ICISTSI-15 , Innovative
Partners for Publishing Solutions, Singapore (May 2015).
Piyush Kumar Pareek & Dr.A.N.Nandakumar, ’Failure Mode Effective Analysis of
Requirements Phase in small software Firms’, Paper ID:
ICSTM/YMCA/2015/292, International Conference on Science, Technology
and Management (ICSTM-2015). International Journal of Advance Research in
Science and Engineering (IJARSE, ISSN- 2319-8354, Impact Factor- 1.142)
[www.ijarse.com], Special Issue Jan2015.
7520

Piyush Kumar Pareek & Dr.A.N.Nandakumar, ’Identifying Wastes in software,

International Journal of Engineering Studies and Technical Approach’.
January Issue 2015.
Piyush Kumar Pareek , Dr.Praveen Gowda , et al ’Ergonomics in a Foundry in
Bangalore to improve productivity’,International Journal of Engineering and
Social Science , ISSN: 2249- 9482 ,Volume 2,Issue 5 (May 2012) , pp 1-6.
Piyush Kumar Pareek , Dr.Praveen Gowda, et al ’FMEA Implementation in a
Foundry in Ban- galore to Improve Quality and Reliability’, International
Journal of Mechanical Engineering and Robotics Research, ISSN :2278-
0149,Volume 1,Issue 2(June 2012),pp 81-87.
Piyush Kumar Pareek et al, ‘Survey on Challenges in Devops ’, International
Journal of Innovative Research in Computer Science & Technology (IJIRCST),
ISSN: 2347-5552, Volume-4, Issue-5, September-2016.
Piyush Kumar Pareek, Dr. A. N. Nandakumar, et al ’Methodology and Functioning
of Project Management Techniques in Agile Software Development Process’,
International Journal of Research in IT, Management and Engineering, ISSN:
2249-1619, Volume2, Issue12 (December2012), pp 76-85.
Piyush Kumar Pareek, Dr. Vasanth Kumar S A , et al ’Reduction of Cycle Time By
Implementation of a Lean Model Carried Out In a Manufacturing Industry’,
International Journal of Engineering and Social Science , ISSN: 2249-
9482,Volume 2, Issue 5, pp 114-123.
Piyush Kumar Pareek, Dr.Vasanth Kumar S A , et al ’Implementation of a Lean
Model for Carrying out Value Stream Mapping in a Manufacturing Industry’,
International Journal of Mechanical Engineering and Robotics Research, ISSN
:2278-0149,Volume 1,Issue 2(June 2012),pp 88-95.
Pooja Chopra, Vijay Suresh Gollamandala, Ahmed Najat Ahmed, S. B. G. Tilak
Babu, Chamandeep Kaur, N. Achyutha Prasad, Stephen Jeswinde
Nuagah, " Automated Registration of Multiangle SAR Images Using
Artificial Intelligence & quot, Mobile Information Systems, vol. 2022, Article
ID 4545139, 10 pages, 2022. https://doi.org/10.1155/2022/4545139.
Prasad N. Achyutha, Sushovan Chaudhury, Subhas Chandra Bose, Rajnish Kler,
Jyoti Surve, Karthikeyan Kaliyaperumal, "User Classification and Stock
Market-Based Recommendation Engine Based on Machine Learning and
Twitter Analysis", Mathematical Problems in Engineering, vol. 2022, Article ID
4644855, 9 pages, 2022. https://doi.org/10.1155/2022/4644855.
R. V S and Siddaraju, "Defective Motes Uncovering and Retrieval for Optimized
Network," 2022 6th International Conference on Computing Methodologies and
Communication (ICCMC), 2022, pp. 303-313, doi:
10.1109/ICCMC53470.2022.9754109.
Rautray, R. and Balabantaray, R. C. (2017). An evolutionary framework for multi
document summarization using cuckoo search approach: Mdscsa. Applied
Computing and Informatics. Salton, G. and McGill, M. J. (1986). Introduction
to modern information retrieval. Sanchez-Gomez, J. M., Vega-Rodr´ıguez, M.
A., and Perez, ´ C. J. (2018). Extractive multi-document text summarization
using a multi-objective artificial bee colony optimization approach. Knowledge-
Based Systems, 159:1–8.
Rekha VS, Siddaraju., “An Ephemeral Analysis on Network Lifetime Improvement
Techniques for Wireless Sensor Networks”, International Journal of Innovative
Technology and Exploring Engineering, vol. 8, issue 9, 2278-3075, pp. 810–
814, 2019.
 7521

Sagar, Y.S. and Achyutha Prasad, N., CHARM: A Cost-Efficient Multi-Cloud Data
Hosting Scheme With High Availability, International Journal for Technological
Research In Engineering, Volume 5, Issue 10, June-2018, ISSN (Online): 2347
– 4718.
Suryasa, I. W., Rodríguez-Gámez, M., & Koldoris, T. (2022). Post-pandemic health
and its sustainability: Educational situation. International Journal of Health
Sciences, 6(1), i-v. https://doi.org/10.53730/ijhs.v6n1.5949
Towards a unified approach to simultaneous single-document and multi-
document summarizations. In Proceedings of the 23rd international conference
on computational linguistics, pages 1137–1145. Association for Computational
Linguistics.
Udit Shinghal, Yashwanth A V Mowdhgalya, Vaibhav Tiwari, Achyutha Prasad
N "Centaur - A Self-Driving Car" International Journal of Computer Trends and
Technology 68.4 (2020):129-131.
Udit Shinghal, Yashwanth A V Mowdhgalya, Vaibhav Tiwari, Achyutha Prasad
N "Home Automation using HTTP and MQTT Server" International Journal
of Computer Trends and Technology 68.4 (2020):126-128.
Verma, P., Pal, S., and Om, H. (2019). A comparative analysis on hindi and
english extractive text summarization. ACM Transactions on Asian and Low-
Resource Language Information Processing (TALLIP), 18(3):30. Wan, X. (2010).
Widyaningrum, I. ., Wibisono, N. ., & Kusumawati, A. H. . (2020). Effect of
extraction method on antimicrobial activity against staphylococcus aureus of
tapak liman (elephantopus scaber l.) leaves. International Journal of Health &
Medical Sciences, 3(1), 105-110. https://doi.org/10.31295/ijhms.v3n1.181

View publication stats