Research paper
saraemily
OriginalityReport.pdf
%67
%1
%1
SafeAssign Originality Report
%69Total Score: High risk Submission UUID: f4c068ff-4928-cd77-e068-bcb8cc87644f
Total Number of Reports
1 Highest Match
69 % Average Match
69 % Submitted on
05/31/20 02:49 PM EDT
Average Word Count
1,332 Highest:
%69Attachment 1
Institutional database (7)
Student paper Student paper Student paper
My paper Student paper Student paper
Student paper
Internet (2)
springer b-ok
Global database (1)
Student paper
Top sources (3)
Excluded sources (0)
Word Count: 1,332
5 2 6
3 1 8
4
10 9
7
5 Student paper 2 Student paper 6 Student paper
RUNNING HEAD: EFFICIENTLY MODEL UNCERTAINTY IN ML AND NLP, AND UNCERTAINTY RESULTING FROM BIG DATA ANALYTICS
Efficiently model uncertainty in ML and NLP, and uncertainty resulting from Big Data Analytics
ITS 836
Data Science & Big Data Analytics
Submitted by
Prof: Dr.
Date: May 31th, 2020
Introduction
When dealing with big data analytics, ML is commonly used to make models for forecast and knowledge detection to empower information-driven dynamics.
Conventional ML techniques are not computationally efficient or adaptable enough to deal with both the attributes of big data (eg, huge volumes, high speeds, shifting sorts, low-value density, inadequacy) and vulnerability (eg, inclined training data, surprising information types, and so forth.). A few usually utilized impelled ML procedures proposed for enormous information examination incorporate element learning, profound learning, move learning, circulated learning, and dynamic
1
2
3
4
5
Source Matches (23)
learning (Ghavami, 2019). Feature learning includes a lot of methods that empower a framework to naturally find the portrayals required for include recognition or classification from unprocessed information. Examining Analytics Techniques
The performances of the ML algorithms are firmly influenced by choice of information depiction. Deep learning algorithms are intended for breaking down and
removing vital information from large measures of data and information gathered from different sources (eg, separate varieties within a picture, for example, a light, different materials, and shapes), nevertheless current deep learning models acquire a high computational expense. Distributed learning can be utilized to
moderate the adaptability issue of customary ML via completing computations on informational indexes appropriated among a few workstations to scale up the learning procedure. Transfer learning is the ability to use information acquired from one setting to a new setting, effectively improving a student from one area by moving data from a related space. Dynamic learning alludes to calculations that utilize versatile information collection (i.e., forms that consequently alter parameters to gather the most helpful information as fast as could reasonably be expected) so as to quicken ML exercises and overcome naming issues (Dasgupta, 2018). The vulnerability difficulties of ML procedures can be basically ascribed to gaining from information with low veracity (i.e., dubious and inadequate information) and information with little value (i.e., irrelevant to the present issue). Among the ML methods, active learning, profound learning, and fluffy rationale hypothesis are extraordinarily fit to help the test of diminishing vulnerability. The vulnerability can affect ML as far as inadequate or uncertain training tests, indistinct classification limits, and harsh information on the objective information. At times, the information is spoken to without names, which can turn into a test. Physically marking enormous information assortments can be a costly and exhausting undertaking yet gaining from unlabeled information is very difficult as characterizing information with hazy rules yields muddled outcomes. Active learning has explained this issue by choosing a subset of the most significant occasions for marking. Profound learning is another learning strategy that can deal with inadequacy and irregularity issues in the classification methodology. NLP has an established set of methodologies, tools, and techniques that cover both written and spoken (not to mention signed) languages (Nasraoui & N'Cir, 2018). Also, it has large application areas such as machine translation, information extraction, speech recognition, optical character recognition, spell checking, and such. Machine Learning (ML), on the other hand, is an approach that could be used in Natural Language Processing and many other fields such as data sciences, decision-making systems, and artificial intelligence. We can perhaps say that NLP is an interdisciplinary field in computing, while ML is a set of approaches and tools to address and solve different problems in a variety of computing fields, including NLP. However, we should not forget that these topics are so getting entangled and intertwined, which makes it difficult to establish a clear line between their definitions. Natural language processing provides clarification to the above-mentioned problems using the vocabulary selection method, understanding synonyms, antonyms, homonyms using wordnet, lexicon formation, relationship identification, and Name entity recognition (Stanford parser). NLP is an aid to ML and also Deep futuristic learning. Moreover, NLP augmented to ML reduce the search space and make it a guided search. As a result, classier don't overfit while training and accuracy are improved. The addition of Semantics to NLP is a major thrust in today's Learning community. NLP is a strategy integrated
into ML that empowers gadgets to assess, decipher, and even create content.
5
2
5
NLP and big data handle large measures of content information and can get an incentive from such a dataset progressively. Some standard NLP practices
include lexical procurement, word sense disambiguation (i.e., figuring out which feeling of the word is utilized in a sentence when a name has different implications), and grammatical feature (POS) labeling (i.e., hinder mining the capacity of the words through marking classes, for example, action word, thing, and so forth). A few NLP-based methods have been used to content mining, including data extraction, theme demonstrating, content outline, classification, grouping, and question feedback, as well as supposition mining. For instance, financial and extortion examinations may include finding proof of wrongdoing in massive datasets (Morabito, 2017). NLP strategies (uniquely named substance extraction and data recovery) can help oversee and filter through colossal measures of literary data, for
example, criminal names and bank records, to support misrepresentation evaluation. Additionally, NLP procedures can assist with making new traceable interfaces and recoup detectability joins by finding semantic closeness among available printed ancient rarities. Moreover, NLP and big data can be utilized to investigate news stories, and foresee rises and falls on the composite stock value file. Vulnerability influences NLP in vast information in various ways. For instance, a catchphrase
search is an exemplary methodology in content mining that is used to deal with a lot of literary knowledge. Watchword search acknowledges as information a rundown of applicable words or expressions and searches the ideal arrangement of data for events of the significant words. The vulnerability can affect
catchphrase search, as an archive that contains a watchword isn't a confirmation of a report's pertinence. For instance, a catchphrase search, for the most part, coordinates accurate strings and overlooks words with a spelling error that may at present be important. Boolean administrators and fluffy pursuit innovations license more prominent flexibility in that they can be utilized to scan for words like the ideal spelling. Conclusion
While big data using AI holds a ton of guarantee, a broad scope of difficulties is presented when such methods are exposed to vulnerability. For example, every
one of the V attributes present various sources of weakness, for example, unstructured, inadequate, or noisy data. Moreover, the vulnerability can be installed in
the whole assessment process. For instance, managing insufficient and lose data is a basic test for most information mining and ML procedures (Ghosh &
Livingston, 2019). Also, an ML algorithm may not get the ideal outcome if the preparation information is one-sided in any capacity. Scaling these worries up to the high data level will effectively exacerbate any errors or inadequacies of the whole investigation process. Accordingly, a moderating vulnerability in big data analytics must be at the cutting edge of any robotized procedure, as vulnerability can have a significant influence on the exactness of its outcomes. 1000 companies that
have had the capability to implement big data analytics with business intelligence. The corporation has used this technology in improving its operations and also in improving the experience of its customers while shopping at its various outlets and online.
References
Dasgupta, N. (2018). Big Data Analytics: Techniques to implement enterprise analytics and machine learning. Birmingham, England: Packt Publishing.
Ghavami, P. (2019). Big data analytics methods: Analytics techniques in data mining, deep learning and natural language processing. de Gruyter. Ghosh, R., &
Livingston, L. J. (2019). Big data analytics: Systems, algorithms, applications. Springer Nature. Morabito, V. (2017). Big Data and Analytics: Strategic and
Organizational Impacts. Basingstoke, England: Springer. Nasraoui, O., & N'Cir, C. B. (2018). Clustering methods for big data analytics: Techniques, toolboxes and
applications. Springer.
5
5
5
6
5
6
5
3
5 7
8
5 9 5
10
Student paper 79%
Student paper 100%
My paper 100%
Student paper 100%
Student paper 71%1
Student paper
EFFICIENTLY MODEL UNCERTAINTY IN ML AND NLP, AND UNCERTAINTY RESULTING FROM BIG DATA ANALYTICS Efficiently model uncertainty in ML and NLP, and uncertainty resulting from Big Data Analytics
Original source
This paper discusses efficiently model uncertainty in ML and NLP, as well as how to represent uncertainty resulting from big data analytics This paper discusses efficiently model uncertainty in ML and NLP, as well as how to represent uncertainty resulting from big data analytics
2
Student paper
Data Science & Big Data Analytics
Original source
Data Science & Big Data Analytics
3
Student paper
Original source
4
Student paper
Original source
5
Student paper
When dealing with big data analytics, ML is commonly used to make models for forecast and knowledge detection to empower information-driven dynamics. Conventional ML techniques are not computationally efficient or adaptable enough to deal with both the attributes of big data (eg, huge volumes, high speeds, shifting sorts, low-value density, inadequacy) and vulnerability (eg, inclined training data, surprising information types, and so forth.). A few usually utilized impelled ML procedures proposed for enormous information examination incorporate element learning, profound learning, move learning, circulated learning, and dynamic learning (Ghavami, 2019). Feature learning includes a lot of methods that empower a framework to naturally find the portrayals required for include recognition or classification from unprocessed information.
Original source
When managing information investigation, ML is commonly used to make models for forecast and information disclosure to empower information-driven dynamics Conventional ML techniques are not computationally effective or versatile enough to deal with both the attributes of enormous details (eg, huge volumes, high speeds, shifting sorts, low worth thickness, deficiency) and vulnerability (eg, one-sided preparing information, unforeseen information types, and so forth.) A few customarily utilized propelled ML systems proposed for tremendous information examination incorporate element learning, profound learning, move education, conveyed learning, and dynamic learning Highlight learning includes a lot of procedures that empower a framework to consequently find the portrayals required for include discovery or grouping from crude information
Student paper 78% Student paper 80%5
Student paper
Deep learning algorithms are intended for breaking down and removing vital information from large measures of data and information gathered from different sources (eg, separate varieties within a picture, for example, a light, different materials, and shapes), nevertheless current deep learning models acquire a high computational expense.
Original source
Profound learning calculations are intended for breaking down and removing important information from large measures of news and data gathered from different sources (eg, separate varieties inside a picture, for example, a light, different materials, and shapes) [Najafabadi MM,2016], anyway current profound learning models bring about a high computational expense
2
Student paper
Distributed learning can be utilized to moderate the adaptability issue of customary ML via completing computations on informational indexes appropriated among a few workstations to scale up the learning procedure. Transfer learning is the ability to use information acquired from one setting to a new setting, effectively improving a student from one area by moving data from a related space. Dynamic learning alludes to calculations that utilize versatile information collection (i.e., forms that consequently alter parameters to gather the most helpful information as fast as could reasonably be expected) so as to quicken ML exercises and overcome naming issues (Dasgupta, 2018). The vulnerability difficulties of ML procedures can be basically ascribed to gaining from information with low veracity (i.e., dubious and inadequate information) and information with little value (i.e., irrelevant to the present issue).
Original source
Disseminated learning can be utilized to moderate the versatility issue of conventional ML via completing computations on informational indexes appropriated among a few workstations to scale up the learning procedure [63] Move learning is the capacity to apply information learned in one setting to new settings, successfully improving a student from one space by moving data from a related area [64] Dynamic learning alludes to calculations that utilize versatile information assortment [65] (i.e., forms that consequently change parameters to gather the most valuable information as fast as could reasonably be expected) so as to quicken ML exercises and defeat marking issues The vulnerability difficulties of ML procedures can be primarily ascribed to gaining from information with low veracity (i.e., questionable and deficient information) and information with low worth (i.e., irrelevant to the present issue)
Student paper 73% Student paper 69%
Student paper 78%
2
Student paper
Among the ML methods, active learning, profound learning, and fluffy rationale hypothesis are extraordinarily fit to help the test of diminishing vulnerability. The vulnerability can affect ML as far as inadequate or uncertain training tests, indistinct classification limits, and harsh information on the objective information.
Original source
We found that, among the ML strategies, dynamic learning, profound learning, and fluffy rationale hypothesis are extraordinarily fit to help the test of lessening vulnerability, as appeared in Fig Vulnerability can affect ML as far as deficient or loose preparing tests, indistinct arrangement limits, and unpleasant information on the objective information
5
Student paper
NLP is a strategy integrated into ML that empowers gadgets to assess, decipher, and even create content. NLP and big data handle large measures of content information and can get an incentive from such a dataset progressively. Some standard NLP practices include lexical procurement, word sense disambiguation (i.e., figuring out which feeling of the word is utilized in a sentence when a name has different implications), and grammatical feature (POS) labeling (i.e., hinder mining the capacity of the words through marking classes, for example, action word, thing, and so forth). A few NLP-based methods have been used to content mining, including data extraction, theme demonstrating, content outline, classification, grouping, and question feedback, as well as supposition mining.
Original source
NLP is a system grounded in ML that empowers gadgets to dissect, decipher, and even produce content [Chen M,2014] NLP and enormous information examination handle large measures of content information and can get an incentive from such a dataset continuously [Wang L,2016] Some conventional NLP techniques incorporate lexical securing (i.e., gets data about the lexical units of a language), word sense disambiguation (i.e., figuring out which feeling of the word is utilized in a sentence when a word has numerous implications), and grammatical form (POS) labeling (i.e., deciding the capacity of the terms through naming classes, for example, action word, thing, and so on.) A few NLP-based strategies have been applied to content mining, including data extraction, theme demonstrating, content synopsis, characterization, bunching, question replying, and supposition mining [Chen M,2014]
5
Student paper
NLP strategies (uniquely named substance extraction and data recovery) can help oversee and filter through colossal measures of literary data, for example, criminal names and bank records, to support misrepresentation evaluation.
Original source
NLP methods (uniquely named element extraction and data recovery) can help oversee and filter through tremendous measures of literary data, for example, criminal names and bank records, to help extortion examinations
Student paper 76%
Student paper 79%
Student paper 68%
Student paper 84%
Student paper 73%
5
Student paper
For instance, a catchphrase search is an exemplary methodology in content mining that is used to deal with a lot of literary knowledge. Watchword search acknowledges as information a rundown of applicable words or expressions and searches the ideal arrangement of data for events of the significant words.
Original source
For instance, a watchword search is an exemplary methodology in content mining that is utilized to deal with a lot of printed information Catchphrase search acknowledges as information a rundown of applicable words or expressions and searches the ideal arrangement of data (eg, a report or database) for events of the related words (i.e., search terms)
6
Student paper
The vulnerability can affect catchphrase search, as an archive that contains a watchword isn't a confirmation of a report's pertinence. For instance, a catchphrase search, for the most part, coordinates accurate strings and overlooks words with a spelling error that may at present be important. Boolean administrators and fluffy pursuit innovations license more prominent flexibility in that they can be utilized to scan for words like the ideal spelling.
Original source
Also, vulnerability can affect catchphrase search, as an archive that contains a watchword isn't an affirmation of a record's significance (Crabb, 2014) For instance, a watchword search for the most part coordinates precise strings and overlooks words with spelling mistakes that may at present be important Boolean administrators and fluffy pursuit innovations license more prominent adaptability in that they can be utilized to look for words like the ideal spelling (Crabb, 2014)
5
Student paper
While big data using AI holds a ton of guarantee, a broad scope of difficulties is presented when such methods are exposed to vulnerability. For example, every one of the V attributes present various sources of weakness, for example, unstructured, inadequate, or noisy data.
Original source
While tremendous information examination utilizing AI holds a great deal of guarantee, a full scope of difficulties is presented when such strategies are exposed to vulnerability Every one of the V attributes present various wellsprings of weakness, for example, unstructured, fragmented, or loud information
6
Student paper
Moreover, the vulnerability can be installed in the whole assessment process.
Original source
Moreover, vulnerability can be installed in the whole examination process
5
Student paper
For instance, managing insufficient and lose data is a basic test for most information mining and ML procedures (Ghosh & Livingston, 2019). Also, an ML algorithm may not get the ideal outcome if the preparation information is one- sided in any capacity. Scaling these worries up to the high data level will effectively exacerbate any errors or inadequacies of the whole investigation process.
Original source
For instance, managing inadequate and uncertain data is a basic test for most information mining and ML procedures Furthermore, an ML calculation may not get the ideal outcome if the preparation information is one-sided in any capacity [Maugis PA,2018] Scaling these worries up to the vast information level will adequately intensify any mistakes or deficiencies of the whole investigation process
My paper 100%
Student paper 100%
Student paper 67%
Student paper 100%
Student paper 100%
b-ok 76%
Student paper 96%
springer 100%
3
Student paper
1000 companies that have had the capability to implement big data analytics with business intelligence. The corporation has used this technology in improving its operations and also in improving the experience of its customers while shopping at its various outlets and online.
Original source
1000 companies that have had the capability to implement big data analytics with business intelligence The corporation has used this technology in improving its operations and also in improving the experience of its customers while shopping at its various outlets and online
5
Student paper
Big Data Analytics:
Original source
Big data analytics
7
Student paper
Techniques to implement enterprise analytics and machine learning.
Original source
Hands-on techniques to implement enterprise analytics and machine learning using Hadoop, Spark, NoSQL and R
8
Student paper
Big data analytics methods: Analytics techniques in data mining, deep learning and natural language processing.
Original source
Big data analytics methods Analytics techniques in data mining, deep learning and natural language processing
5
Student paper
Big data analytics:
Original source
Big data analytics
9
Student paper
Systems, algorithms, applications.
Original source
Algorithms and Applications
5
Student paper
Big Data and Analytics:
Original source
Big data analytics
10
Student paper
Clustering methods for big data analytics: Techniques, toolboxes and applications.
Original source
Clustering Methods for Big Data Analytics Techniques, Toolboxes and Applications
