-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathelements.html
259 lines (202 loc) · 16.7 KB
/
elements.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
<!DOCTYPE HTML>
<!--
Industrious by TEMPLATED
templated.co @templatedco
Released for free under the Creative Commons Attribution 3.0 license (templated.co/license)
-->
<html>
<head>
<title>Ashok Ajad | Machine Learning Engineer</title>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1, user-scalable=no" />
<meta name="description" content="" />
<meta name="keywords" content="" />
<link rel="stylesheet" href="assets/css/main.css" />
</head>
<body class="is-preload">
<!-- Header -->
<header id="header">
<div class="inner">
<div class="content">
<section>
<h4></h4>
<ul class="plain">
<li><a href="https://www.linkedin.com/in/ashok-ajad/" target="_blank"><i class="icon fa-linkedin"> </i>LinkedIn</a>
<a href="https://github.com/code-cse" target="_blank"><i class="icon fa-github"> </i>Github</a>
</li>
</ul>
</section>
</div>
</div>
<a class="logo" href="index.html">Home</a>
<a class="logo" href="index.html">Blogs</a>
<a class="logo" href="generic.html">Research</a>
<a class="logo" href="resume.html">Resume</a>
<nav>
<a href="#menu">Menu</a>
</nav>
</header>
<!-- Nav -->
<nav id="menu">
<ul class="links">
<li><a href="index.html">Home</a></li>
<li><a href="elements.html">Projects</a></li>
<li><a href="generic.html - Details">Research</a></li>
<li><a href="generic.html">Blogs</a></li>
<li><a href="resume.html">Resume</a></li>
</ul>
</nav>
<!-- Heading -->
<div id="heading" >
<h1>Projects</h1>
</div>
<section id="main" class="wrapper">
<div class="inner">
<div class="content">
<!-- <header>
<h2>Comparative study of Object Detection</h2>
</header> -->
<h3><b style="font-size: 20px">Projects - Details</b></h3><p><b style="font-size: 16px;">Auto Response Deal </b><br> Problem Statement: To generate auto-response samples to assist agents in responding faster & better to incoming customer emails based on the contents of the emails using NLP powered solution. NLP powered solution for 1st level of auto-reply that generates 3 sample responses for the agents to choose from to reply. This solution is set up and integrated into the zendesk environment. The solution has an active learning pipeline packaged within the solution.
</br>
Output: Generate top 3 email response for the customer query and assigned to the best agent related to query.
</br>
<b style="font-size: 12px; color:#0000FF">Research paper used in AI-Model.</b>
</br>
<b>1. BERT | <a style="color: #0000FF">Source:</a> <a href="https://arxiv.org/abs/1810.04805"; target="_blank"> BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding </a></b><br>
<b>2. OpenAI-GPT2 | <a style="color: #0000FF">Source:</a> <a href="https://openai.com/blog/better-language-models/"; target="_blank">Language Models are Unsupervised Multitask Learners</a></b><br>
<b>3. Smart Reply: ARS-Email | <a style="color: #0000FF">Source:</a> <a href="https://www.kdd.org/kdd2016/papers/files/Paper_1069.pdf"; target="_blank">Smart Reply: Automated Response Suggestion for Email</a></b><br>
</p>
<p><b style="font-size: 16px;">Healthcare Domain : Multiple usecase related to Radiology and Ophthalmology department. </b><br> Problem Statement:To build a custom model for multiple use cases of Radiology and Ophthalmology.
</br>
<b style="font-size: 12px; color:#0000FF">Research paper used in AI-Model.</b>
</br>
<!-- <b>1. Object Detection | <a style="color: #0000FF">Source:</a> <a href="https://arxiv.org/pdf/1512.02325.pdf"; target="_blank"> SSD: Single Shot MultiBox Detector</a></b><br>
<b>2. Face-Recognition | <a style="color: #0000FF">Source:</a> <a href="https://arxiv.org/abs/1503.03832"; target="_blank">FaceNet: A Unified Embedding for Face Recognition and Clustering</a></b><br>
<b>3. Face-Tracking | <a style="color: #0000FF">Source:</a> <a href="https://pdfs.semanticscholar.org/8461/7541ad311942be796095cb54e970c578307c.pdf"; target="_blank">An Algorithm For Centroid-Based Tracking of Moving Objects </a></b><br>
-->
</p>
<p><b style="font-size: 16px;">Monitor the Person Appearance </b><br> Problem Statement: Monitor the person’s attendance. Associate the faces with the existing employee database to identify each employee record the time stamp and date for each employee recognized.
</br>
Data-set : 100 hrs. Video footage for analysis and to build out model for monitor of person appearance.
</br>
Output: The output of the analysis will be recorded in an csv file, which will record the person’s id and the his/her time stamp and date
</br>
<b style="font-size: 12px; color:#0000FF">Research paper used in AI-Model.</b>
</br>
<b>1. Object Detection | <a style="color: #0000FF">Source:</a> <a href="https://arxiv.org/pdf/1512.02325.pdf"; target="_blank"> SSD: Single Shot MultiBox Detector</a></b><br>
<b>2. Face-Recognition | <a style="color: #0000FF">Source:</a> <a href="https://arxiv.org/abs/1503.03832"; target="_blank">FaceNet: A Unified Embedding for Face Recognition and Clustering</a></b><br>
<b>3. Face-Tracking | <a style="color: #0000FF">Source:</a> <a href="https://pdfs.semanticscholar.org/8461/7541ad311942be796095cb54e970c578307c.pdf"; target="_blank">An Algorithm For Centroid-Based Tracking of Moving Objects </a></b><br>
</p>
<p><b style="font-size: 16px;">Monitor the Person Appearance </b><br> Problem Statement: Monitor the person’s attendance. Associate the faces with the existing employee database to identify each employee record the time stamp and date for each employee recognized.
</br>
Data-set : 100 hrs. Video footage for analysis and to build out model for monitor of person appearance.
</br>
Output: The output of the analysis will be recorded in an csv file, which will record the person’s id and the his/her time stamp and date
</br>
<b style="font-size: 12px; color:#0000FF">Research paper used in AI-Model.</b>
</br>
<b>1. Object Detection | <a style="color: #0000FF">Source:</a> <a href="https://arxiv.org/pdf/1512.02325.pdf"; target="_blank"> SSD: Single Shot MultiBox Detector</a></b><br>
<b>2. Face-Recognition | <a style="color: #0000FF">Source:</a> <a href="https://arxiv.org/abs/1503.03832"; target="_blank">FaceNet: A Unified Embedding for Face Recognition and Clustering</a></b><br>
<b>3. Face-Tracking | <a style="color: #0000FF">Source:</a> <a href="https://pdfs.semanticscholar.org/8461/7541ad311942be796095cb54e970c578307c.pdf"; target="_blank">An Algorithm For Centroid-Based Tracking of Moving Objects </a></b><br>
</p>
<hr />
<p><b style="font-size: 16px;">DNA-Splice Gene Prediction</b><br> Each DNA read is a sequence of four [C,A,G,T] types of nucleotides and needs to be converted into numerical representations for machine learning.
The domain consists of 60 variables, representing a sequence of DNA bases an additional class Variable. The task is to determine if the middle of the sequence is a splice junction and what is its type: Splice junctions are of two types:
</br>
1. exon-intron (EI): represents the end of an exon and the beginning of an intron
</br>
2. intron-exon (IE): represents where the intron ends and the next exon, or coding section, begins.
</br>
So the class variable contains 3 values:
</br>
1. exon-intron (EI) | 2. intron-exon (IE) | 3. No-Junction
</br>
Other 4 values corresponding to the 4 possible DNA bases (C, A, G, T)
</br>
C : Cytosine | A : Adenine | G : Guanine | T : Thymine
</br>
<b style="font-size: 12px; color:#0000FF">Research paper used in AI-Model.</b>
</br>
<b>1. DNA-Splice-Jucntion | <a style="color: #0000FF">Source:</a> <a href="https://arxiv.org/pdf/1512.05135.pdf"; target="_blank"> DNA-Level Splice Junction Prediction using Deep Recurrent Neural Networks</a></b><br>
<b>2. Classification of DNA-Splice | <a style="color: #0000FF">Source:</a> <a href="https://core.ac.uk/download/pdf/82482313.pdf"; target="_blank">Classification of eukaryotic splice-junction genetic sequences using averaged one-dependence estimators with subsumption resolution</a></b><br>
</p>
<hr />
<p><b style="font-size: 16px;">Quality Enhancement of Images Using GAN </b><br>
Problem of this project is that we have to build a model which makes it possible to generate HR-Images based on their LR-Images
</br>
A deep-learning-based solution for the construction of a super resolution images. Trained the model on 200k LR-Images and the condition to be enhanced of LR-Images.
</br>
<b style="font-size: 12px; color:#0000FF">Research paper used in AI-Model.</b>
</br>
<b>1. SR-GAN | <a style="color: #0000FF">Source:</a> <a href="https://arxiv.org/abs/1609.04802"; target="_blank"> Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network</a></b><br>
<b>2. Condition SR-GAN | <a style="color: #0000FF">Source:</a> <a href="http://cs231n.stanford.edu/reports/2017/pdfs/314.pdf"; target="_blank">Class-Conditional Superresolution with GANs</a></b><br>
</p>
<hr />
<p><b style="font-size: 16px;">Data Warehouse System </b><br> The problem of this project is that we have to design a scalable data warehouse system that contains various categories and classified image in that categories with no duplication and various different operation. Trained the model on more than 100k images for our data warehouse system and perform different operation on these images. Integrate the whole system in the pipeline and create a GUI and API for the user interface. The accuracy of the system is 99.38 %.
</br>
<b style="font-size: 12px; color:#0000FF">Research paper used in AI-Model.</b>
</br>
<b>1. VGG-16 Architecture | <a style="color: #0000FF">Source:</a> <a href="https://arxiv.org/abs/1409.1556"; target="_blank"> Very Deep Convolutional Networks for Large-Scale Image Recognition</a></b><br>
<b>2.Hashing Techniques | <a style="color: #0000FF">Source:</a> <a href="https://core.ac.uk/download/pdf/35471499.pdf"; target="_blank">Selecting a Hashing Algorithm</a></b><br>
<b>2. Simialrity Algorithm | <a style="color: #0000FF">Source:</a> <a href="http://ai.stanford.edu/~gal/Papers/chechik_nips2009.pdf"; target="_blank">An Online Algorithm for Large Scale Image Similarity Learning</a></b><br>
</p>
<hr />
<p><b style="font-size: 16px;">Recommender System</b><br> The problem of this project is that we have to design a recommender system for male & female based on their dressing style and face shape etc. For this project, we have used the detection and classification model. Trained the model on the huge number of label images for our detection and classification model and perform different operation on these detected images. Integrate the whole system in the pipeline and create a GUI and API for the user interface.<br>
<b style="font-size: 12px; color:#0000FF">Research paper used in AI-Model.</b>
</br>
<b>1. Object Detection | <a style="color: #0000FF">Source:</a> <a href="https://arxiv.org/abs/1506.02640"; target="_blank"> You Only Look Once: Unified, Real-Time Object Detection</a></b><br>
<b>2. Classification | <a style="color: #0000FF">Source:</a> <a href="https://arxiv.org/abs/1512.00567"; target="_blank">Rethinking the Inception Architecture for Computer Vision</a></b><br>
<b>3. Clustering | <a style="color: #0000FF">Source:</a> <a href="https://arxiv.org/pdf/1002.2425.pdf"; target="_blank">Application of k-Means Clustering algorithm for prediction of Students’ Academic Performance</a></b><br>
<b>4. Recommender System (using Bi-Directional LSTM) | <a style="color: #0000FF">Source:</a> <a href="https://static.googleusercontent.com/media/research.google.com/en//pubs/archive/43905.pdf"; target="_blank">Long Short-Term Memory Recurrent Neural Network Architectures for Large Scale Acoustic Modeling</a></b><br>
</p>
<hr />
<p><b style="font-size: 16px;">Chatbot for Automatic Question Answer</b><br>The problem of this project is that we have to build a chatbot which makes it possible to generate responses based on collections of known conversations. A deep-learning-based solution for the construction of a chatbot off-topic behavior and personality. When confronted with off-topic questions, the bot will try to automatically generate a possibly relevant answer from scratch, based only on a pre-trained RNN model.
</p>
<hr />
<h3><b style="font-size: 20px">Course Projects</b></h3>
<p><b style="font-size: 16px;">Face Shape Type Detector of Human Being</b><br> Implemented a system that analyses various type of face like square, oblong, oval, heart, round and diamond. Dataset gathered from various internet resources and clean the data according to our requirements. Trained the model on more than 10000 labeled images for face shape type detection. The accuracy of system is 84.56%.
</p>
<hr />
<p><b style="font-size: 16px;">Keyword Extraction and Semantic Tag Prediction [KESTP]</b><br>APPLIED SUPERVISED LEARNING METHODS FOR PREDICTING TAGS <br>• Implemented a system that analyses the tags for a question asked by user into appropriate categories.<br>• Data set of Stack overflow dump is used to predict the tags according to user questions.<br>• Applied java sax parser to parse the useful data from data dump.<br>• Applied different classifiers on title ,body and code to predicts the tags.<br>• Used M.L algorithms like SVM , Naive Bayes & Unique Feature Extraction to predict tags.<br>
</p>
<hr />
<p><b style="font-size: 16px;">House Price Prediction-Regression Techniques [HPPRT]</b><br> APPLIED MACHINE LEARNING ALGORITHM <br> • Devised a mechanism to predict the price of each house using stacking of various outputs of regression techniques like lasso and ridge regression, XGboost, elastic net.<br> • This Project is implemented as the part of kaggle competition.
</p>
<hr />
<p><b style="font-size: 16px;">Stock Price Trend Forecasting [SPTF]</b><br>APPLIED SUPERVISED LEARNING METHODS FOR STOCK PRICE TREND FORECASTING <br> • Implemented a system that analyses previous stock data of various <br>companies, processes time-series data and aims to forecast the trends of stock in near future.<br>• Data set of Indian stock market is used to predict the next-minute, next day and next week stock trend.<br>• Applied moving averages,Log normalization techniques for feature generation & selection process.<br>• Used M.L algorithms like SVM & Neural network to predict trend.<br>
</p>
<hr />
<p><b style="font-size: 16px;">Search Engine for 81 GB English Wikipedia Dump</b><br> NFORMATION RETRIEVAL <br>• Implemented inverted index and calculated TF-IDF score.<br>• Retrieving the top 10 results for any query search within 0.3-0.6 seconds.<br>• Supports field search on title, infobox, text, categories.
</p>
</div>
</div>
</section>
<!-- *********************************************************************************************************************************** -->
<!-- Footer -->
<footer id="footer">
<div class="inner">
<div class="content">
<section>
<h4></h4>
<ul class="plain">
<li><a href="https://www.linkedin.com/in/ashok-ajad/" target="_blank"><i class="icon fa-linkedin"> </i>LinkedIn</a>
<a href="https://github.com/code-cse" target="_blank"><i class="icon fa-github"> </i>Github</a>
<a href="#"><i class="icon fa-facebook"> </i>Facebook</a>
<a href="#"><i class="icon fa-instagram"> </i>Instagram</a>
</li>
</ul>
</section>
</div>
<div class="copyright">
© ashok ajad <a href="index.html">Machine Learning Engineer | AI-Researcher </a>
</div>
</div>
</footer>
<!-- Scripts -->
<script src="assets/js/jquery.min.js"></script>
<script src="assets/js/browser.min.js"></script>
<script src="assets/js/breakpoints.min.js"></script>
<script src="assets/js/util.js"></script>
<script src="assets/js/main.js"></script>
</body>
</html>