Pub Date : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9909500
Ravi Kiran Mallidi, Manmohan Sharma, Sreenivas Rao Vangala
Rapid development in information technology leads to modernizing the applications with innovations and technologies to achieve time to market, meet customer needs, better UI, security, and efficient system. Several modernization options are present in the market to achieve the above needs in Banking and Financial applications. Banking and Financial application are critical and adopt the newer technologies quickly to achieve performance and security. Nowadays, the end-user is using the technologies at their fingertips. BFSI (Banking, Financial Services and Insurance) enterprises widely spread and interact with each other. Enterprise application integration plays a crucial role in the BFSI domain to achieve technical and business benefits. Kafka is one of the technical solutions to build a real-time streaming platform between applications on moving data from one system to another. Streaming applications are scalable, fast, and durable. Streaming platforms are credit and different from Message Queues for advantages like Scalability, Message retention, multiple consumers, replication, message order, and TCP protocol support. Banking and financial application backend applications, especially batch processing systems, migrate from batch to stream platform to achieve near real-time data. As a result, the processing/settlement processing time is minimized in banking and settlement systems. This paper discusses two case studies and identifies the advantages and disadvantages of streaming platforms.
{"title":"Streaming Platform Implementation in Banking and Financial Systems","authors":"Ravi Kiran Mallidi, Manmohan Sharma, Sreenivas Rao Vangala","doi":"10.1109/ASIANCON55314.2022.9909500","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9909500","url":null,"abstract":"Rapid development in information technology leads to modernizing the applications with innovations and technologies to achieve time to market, meet customer needs, better UI, security, and efficient system. Several modernization options are present in the market to achieve the above needs in Banking and Financial applications. Banking and Financial application are critical and adopt the newer technologies quickly to achieve performance and security. Nowadays, the end-user is using the technologies at their fingertips. BFSI (Banking, Financial Services and Insurance) enterprises widely spread and interact with each other. Enterprise application integration plays a crucial role in the BFSI domain to achieve technical and business benefits. Kafka is one of the technical solutions to build a real-time streaming platform between applications on moving data from one system to another. Streaming applications are scalable, fast, and durable. Streaming platforms are credit and different from Message Queues for advantages like Scalability, Message retention, multiple consumers, replication, message order, and TCP protocol support. Banking and financial application backend applications, especially batch processing systems, migrate from batch to stream platform to achieve near real-time data. As a result, the processing/settlement processing time is minimized in banking and settlement systems. This paper discusses two case studies and identifies the advantages and disadvantages of streaming platforms.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121810992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The following article presents an inner round architecture for the AES Encryption Scheme suitable for implementation on FPGAs and as ASICs. The uniformity between the encryption and decryption hardware makes them suitable for implementation as separate or co-existing blocks as required. The modular approach of our architecture allows for different encryption/decryption core configurations providing a compact, scalable implementation that is suitable for applications that may demand compact yet high performant hardware. The architecture employs a combinational S-Box forming a crucial step in the parallel operation of the hardware. For an operating frequency of 278.5 MHz, the hardware achieves a high throughput of about 3.5 gigabits per second (GBps).
{"title":"An Inner Round Pipeline Architecture Hardware Core for AES","authors":"Archit Jain, Divyanshu Jain, Arpan Katiyar, Gurjit Kaur","doi":"10.1109/ASIANCON55314.2022.9909114","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9909114","url":null,"abstract":"The following article presents an inner round architecture for the AES Encryption Scheme suitable for implementation on FPGAs and as ASICs. The uniformity between the encryption and decryption hardware makes them suitable for implementation as separate or co-existing blocks as required. The modular approach of our architecture allows for different encryption/decryption core configurations providing a compact, scalable implementation that is suitable for applications that may demand compact yet high performant hardware. The architecture employs a combinational S-Box forming a crucial step in the parallel operation of the hardware. For an operating frequency of 278.5 MHz, the hardware achieves a high throughput of about 3.5 gigabits per second (GBps).","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122328044","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9908676
Jayama Pinnamaneni, N. S, Prasad B. Honnavalli
A Docker container image can be defined as a lightweight, unattached, executable package of software that includes everything like code, runtime, system tools, system libraries and settings, needed to run an application, because of these features the container images are preferred over virtual machines. With this enormous usage, there is a lot of scope for the security issues arising in the container images. There are many open-source projects like Anchore, Clair that statically scan the container image’s docker file to find the vulnerabilities using databases like CVE, RedHat etc. Static analysis of container image main code is equally necessary to identify any vulnerabilities in the code and not only focus on the vulnerabilities based on OS level, as many malicious activities might take place if code is not scanned for any vulnerabilities. The main aim of the project is to create a static code analysing machine learning model to identify the vulnerable python libraries in container images.
{"title":"Identifying Vulnerabilities in Docker Image Code using ML Techniques","authors":"Jayama Pinnamaneni, N. S, Prasad B. Honnavalli","doi":"10.1109/ASIANCON55314.2022.9908676","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9908676","url":null,"abstract":"A Docker container image can be defined as a lightweight, unattached, executable package of software that includes everything like code, runtime, system tools, system libraries and settings, needed to run an application, because of these features the container images are preferred over virtual machines. With this enormous usage, there is a lot of scope for the security issues arising in the container images. There are many open-source projects like Anchore, Clair that statically scan the container image’s docker file to find the vulnerabilities using databases like CVE, RedHat etc. Static analysis of container image main code is equally necessary to identify any vulnerabilities in the code and not only focus on the vulnerabilities based on OS level, as many malicious activities might take place if code is not scanned for any vulnerabilities. The main aim of the project is to create a static code analysing machine learning model to identify the vulnerable python libraries in container images.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127010420","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9908772
Vedha Krishna Yarasuri, Dhumsapuram Saikrishna Reddy, Puligundla Sai Muneesh, Ramabhotla Venkata Sai Kaushik, Thupalli Nanda Vardhan, K. L. Nisha
Cardiovascular diseases (CVDs) are a range of heart and blood vessel problems leading to death worldwide. It is critical to discover cardiac diseases as early as feasible in order to extend one's life expectancy. Machine learning is an efficacious method for predicting the presence of severe diseases and the risk they cause to patients. In this paper, five machine learning algorithms namely Logistic Regression, Random Forests, K-Nearest Neighbor, Decision Trees, and Support Vector Machines were executed to predict the risk of cardiovascular diseases. These results can then be used to assist the doctors in identifying the patients with a higher risk of heart failure to ensure timely treatment.
{"title":"Developing Machine Learning Models for Cardiovascular Disease Prediction","authors":"Vedha Krishna Yarasuri, Dhumsapuram Saikrishna Reddy, Puligundla Sai Muneesh, Ramabhotla Venkata Sai Kaushik, Thupalli Nanda Vardhan, K. L. Nisha","doi":"10.1109/ASIANCON55314.2022.9908772","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9908772","url":null,"abstract":"Cardiovascular diseases (CVDs) are a range of heart and blood vessel problems leading to death worldwide. It is critical to discover cardiac diseases as early as feasible in order to extend one's life expectancy. Machine learning is an efficacious method for predicting the presence of severe diseases and the risk they cause to patients. In this paper, five machine learning algorithms namely Logistic Regression, Random Forests, K-Nearest Neighbor, Decision Trees, and Support Vector Machines were executed to predict the risk of cardiovascular diseases. These results can then be used to assist the doctors in identifying the patients with a higher risk of heart failure to ensure timely treatment.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126577310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9908919
Abhinav Rampeesa, Ponnaboina Akhila, Mohammed Irfan, S. Rebelli, L. R. Thoutam, J. Ajayan
The majority of the most recent data-driven advancements, notably artificial-intelligence(AI) and machine-learning (ML) depend heavily on binary arithmetic computations. This paper focuses on the design and analysis of a reliable, low-power 4-bit Baugh-Wooley (BW) multiplier employing the high performance 1-bit mirror full-adder (MFA) and approximate full-adder (AFA). For a various technological nodes varying from 16 nm to 90 nm, the effectiveness of the proposed 4-bit BW multiplier is thoroughly investigated for power and delay parameters at different operating voltages (0.6 V – 1.0 V). The proposed 4-bit BW multiplier at a 16-nm CMOS process employing an MFA circuit consumes a minimal power of 37.5 μW at an operating voltage of 0.7 V and a nominal temperature of 300C, whereas it consumes 35.5 μW for a combined MFA and AFA circuits. The power consumption increases linearly with operating voltage for the designed BW multiplier employing the two high performance full-adder circuits. The proposed 4-bit BW multiplier at 16-nm CMOS process has a delay of 104 ps at 0.7 V. The simulation result analysis indicates the combination of MFA and AFA circuits in the design of low-power 4-bit Baugh-Wooley multipliers, even though there exists a partial error at the output of multiplier MSB.
{"title":"Design of Low Power 4-Bit Baugh-Wooley Multiplier using 1-Bit Mirror and Approximate Full Adders","authors":"Abhinav Rampeesa, Ponnaboina Akhila, Mohammed Irfan, S. Rebelli, L. R. Thoutam, J. Ajayan","doi":"10.1109/ASIANCON55314.2022.9908919","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9908919","url":null,"abstract":"The majority of the most recent data-driven advancements, notably artificial-intelligence(AI) and machine-learning (ML) depend heavily on binary arithmetic computations. This paper focuses on the design and analysis of a reliable, low-power 4-bit Baugh-Wooley (BW) multiplier employing the high performance 1-bit mirror full-adder (MFA) and approximate full-adder (AFA). For a various technological nodes varying from 16 nm to 90 nm, the effectiveness of the proposed 4-bit BW multiplier is thoroughly investigated for power and delay parameters at different operating voltages (0.6 V – 1.0 V). The proposed 4-bit BW multiplier at a 16-nm CMOS process employing an MFA circuit consumes a minimal power of 37.5 μW at an operating voltage of 0.7 V and a nominal temperature of 300C, whereas it consumes 35.5 μW for a combined MFA and AFA circuits. The power consumption increases linearly with operating voltage for the designed BW multiplier employing the two high performance full-adder circuits. The proposed 4-bit BW multiplier at 16-nm CMOS process has a delay of 104 ps at 0.7 V. The simulation result analysis indicates the combination of MFA and AFA circuits in the design of low-power 4-bit Baugh-Wooley multipliers, even though there exists a partial error at the output of multiplier MSB.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124387859","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9908712
Keerthana P.B., Joseph K.D.
Recently, wind energy generation grows quickly because of its economical features and it has less effect on mother earth. The long-distance between generation and customer reduces the maximum transmittable power. For addressing this issue series compensation is broadly used to raise the capacity of transmission. But the insertion of capacitors has the hazardous issue of Sub Synchronous Resonance (SSR). An Enhanced Detection Technique(EDT) is used in Double Fed Induction Generator (DFIG) based wind power system connected to the series compensated line(SCL) for fast detection of SSR. Comparison of enhanced detection technique with the traditional technique validate the superiority of EDT. SSR Damping Controller (SSRDC) in the static synchronous compensator (STATCOM) is applied for mitigation of hazardous effect of SSR. The voltage signal is the input for the detection circuit and line current signal is the input for SSRDC.
{"title":"Enhanced Detection and Mitigation on Sub Synchronous Resonance in Wind Farm with Series Compensated Line","authors":"Keerthana P.B., Joseph K.D.","doi":"10.1109/ASIANCON55314.2022.9908712","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9908712","url":null,"abstract":"Recently, wind energy generation grows quickly because of its economical features and it has less effect on mother earth. The long-distance between generation and customer reduces the maximum transmittable power. For addressing this issue series compensation is broadly used to raise the capacity of transmission. But the insertion of capacitors has the hazardous issue of Sub Synchronous Resonance (SSR). An Enhanced Detection Technique(EDT) is used in Double Fed Induction Generator (DFIG) based wind power system connected to the series compensated line(SCL) for fast detection of SSR. Comparison of enhanced detection technique with the traditional technique validate the superiority of EDT. SSR Damping Controller (SSRDC) in the static synchronous compensator (STATCOM) is applied for mitigation of hazardous effect of SSR. The voltage signal is the input for the detection circuit and line current signal is the input for SSRDC.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128735589","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9909395
Rashi Agarwal, Silky Goel, Rahul Nijhawan
The land cover is the apparent (bio)physical cover, and land use alludes to how the actual land type is being utilized. This research is fundamental to survey the degree to which social, monetary, and natural factors influence urbanization. This will likewise assist with urban planning. As laborious process of handcrafted feature extraction has not helped obtain high accuracies, this paper proposes use of Deep Learning approach that explores different Image Recognition Models using various ML classifiers on remote sensing images classifying the images from large Landsat satellite dataset into 9 different classes. It was observed that the highest accuracy of 97.4% was achieved by the Logistic Regression algorithm coupled with Inceptionv3 model. The proposed model shows the capability of increasing the accuracy of existing state-of-art-algorithms low resolution land classification maps. Thus, the improved results will contribute to better land maps helping with the growing demand of LULC information concerning climate change and sustainable development.
{"title":"Using Deep Learning Approach for Land-Use and Land-Cover Classification based on Satellite images","authors":"Rashi Agarwal, Silky Goel, Rahul Nijhawan","doi":"10.1109/ASIANCON55314.2022.9909395","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9909395","url":null,"abstract":"The land cover is the apparent (bio)physical cover, and land use alludes to how the actual land type is being utilized. This research is fundamental to survey the degree to which social, monetary, and natural factors influence urbanization. This will likewise assist with urban planning. As laborious process of handcrafted feature extraction has not helped obtain high accuracies, this paper proposes use of Deep Learning approach that explores different Image Recognition Models using various ML classifiers on remote sensing images classifying the images from large Landsat satellite dataset into 9 different classes. It was observed that the highest accuracy of 97.4% was achieved by the Logistic Regression algorithm coupled with Inceptionv3 model. The proposed model shows the capability of increasing the accuracy of existing state-of-art-algorithms low resolution land classification maps. Thus, the improved results will contribute to better land maps helping with the growing demand of LULC information concerning climate change and sustainable development.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129189782","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9909322
Nitesh Soni, M. Barai
The main barrier to the widespread adoption of electric vehicles is the low mileage on one charge. A lot of kinetic energy gets wasted on the wheels of the vehicle in the form of heat during braking. Regenerative braking is the method of recovering the kinetic energy from the motor during braking. This paper presents the study of regenerative braking of BLDC motor targeting electric vehicle (EV) applications. The induced back electromotive force (EMF) during braking at the motor terminal is used as a source to recharge the battery. This method of regeneration improves the mileage of an EV and reduces braking time as well. However, the battery can be charged with this back EMF if its magnitude is higher than the battery voltage. A boosting action is performed to boost up the level of this back EMF without using any dedicated DC-DC boost converter or an ultra-capacitor. A three phase two level VSI in the closed loop with BLDC motor load is designed and implemented in MATLAB/Simulink environment. The generations of control signals for two level VSI with BLDC motor in closed loop operation are carried out to perform trapezoidal commutation. The energy recovery operation is verified by charging the battery during the braking of BLDC Motor. Simulation results are presented to illustrate the regenerative braking of the BLDC motor.
{"title":"Performance Study of Regenerative Braking of BLDC Motor targeting Electric Vehicle Applications","authors":"Nitesh Soni, M. Barai","doi":"10.1109/ASIANCON55314.2022.9909322","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9909322","url":null,"abstract":"The main barrier to the widespread adoption of electric vehicles is the low mileage on one charge. A lot of kinetic energy gets wasted on the wheels of the vehicle in the form of heat during braking. Regenerative braking is the method of recovering the kinetic energy from the motor during braking. This paper presents the study of regenerative braking of BLDC motor targeting electric vehicle (EV) applications. The induced back electromotive force (EMF) during braking at the motor terminal is used as a source to recharge the battery. This method of regeneration improves the mileage of an EV and reduces braking time as well. However, the battery can be charged with this back EMF if its magnitude is higher than the battery voltage. A boosting action is performed to boost up the level of this back EMF without using any dedicated DC-DC boost converter or an ultra-capacitor. A three phase two level VSI in the closed loop with BLDC motor load is designed and implemented in MATLAB/Simulink environment. The generations of control signals for two level VSI with BLDC motor in closed loop operation are carried out to perform trapezoidal commutation. The energy recovery operation is verified by charging the battery during the braking of BLDC Motor. Simulation results are presented to illustrate the regenerative braking of the BLDC motor.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132280974","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This work describes a voice recognition system that does not need an intermediate phonetic representation to convert audio input to text. The system is based on a mix of the the Connectionist Temporal Classification goal function and deep bidirectional LSTM recurrent neural network architecture . A new method is proposed in which the network is taught to reduce the likelihood of an arbitrary transcription loss function being encountered. without the aid of any lexicons or models, this allows for a direct optimization of WER. The system has a WER (word error rate) of 22 percent, 20 percent with simply a lexicon of authorized terms, 9 percent using a trigram language model. The error rate drops to 7 percent when the network is used in conjunction with a baseline system.
{"title":"Reinforcement Learning for Speech Recognition using Recurrent Neural Networks","authors":"Imad Burhan Kadhim, Mahdi Fadil Khaleel, Zuhair Shakor Mahmood, Ali Nasret Najdet Coran","doi":"10.1109/ASIANCON55314.2022.9908930","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9908930","url":null,"abstract":"This work describes a voice recognition system that does not need an intermediate phonetic representation to convert audio input to text. The system is based on a mix of the the Connectionist Temporal Classification goal function and deep bidirectional LSTM recurrent neural network architecture . A new method is proposed in which the network is taught to reduce the likelihood of an arbitrary transcription loss function being encountered. without the aid of any lexicons or models, this allows for a direct optimization of WER. The system has a WER (word error rate) of 22 percent, 20 percent with simply a lexicon of authorized terms, 9 percent using a trigram language model. The error rate drops to 7 percent when the network is used in conjunction with a baseline system.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133644808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Missing values in a dataset has always been a problem for data analysis and modelling. Building a model over a dataset where the missing values are not handled properly will definitely degrade the accuracy and performance of model. This problem particularly impacts deterministic models. Knowing that majority of the models that are used today are deterministic makes dealing with missing values crucial before applying the machine learning model. In this paper we have discussed various approaches such as statistical method (using mean), MICE and KNN for imputing missing values and tested their accuracy in combination with two prediction algorithms linear regression and random forest regression. We have used dataset of used cars containing missing values in few columns to predict the price of car given the details of car and thus comparing the accuracy of the estimated price with different approaches.
{"title":"Imputing missing values for Dataset of Used Cars","authors":"Samveg Shah, Mayur Telrandhe, Prathmesh Waghmode, Sunil Ghane","doi":"10.1109/ASIANCON55314.2022.9908600","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9908600","url":null,"abstract":"Missing values in a dataset has always been a problem for data analysis and modelling. Building a model over a dataset where the missing values are not handled properly will definitely degrade the accuracy and performance of model. This problem particularly impacts deterministic models. Knowing that majority of the models that are used today are deterministic makes dealing with missing values crucial before applying the machine learning model. In this paper we have discussed various approaches such as statistical method (using mean), MICE and KNN for imputing missing values and tested their accuracy in combination with two prediction algorithms linear regression and random forest regression. We have used dataset of used cars containing missing values in few columns to predict the price of car given the details of car and thus comparing the accuracy of the estimated price with different approaches.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"122 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131160341","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：