Search engines have incomparable economic values which make all the
search engine tools provider improve their search engine constantly and evolve with the
dynamic nature of data and information need of the user. The relevance of the result
which we get from a specific search engine is user specific and with user satisfaction
which is difficult to measure. To come up with good measures we need to evaluate the
measure on the different set of queries with different relevance and information need.
Buckley and Voorhees (2000) compare several evaluation measures and discuss the way to
create the Training dataset and Test dataset and the stability of different measures
based of varying search engine parameters like a number of query sets and the number of
documents needed. On similar lines, we have developed the search engine on training
dataset which has both Boolean and Ranked retrieval capabilities and four different
ranking algorithms.
Techniques Implemented:
- tf-idf (term frequency)
- The weight of the document in the corpus
- The weight of the term in a document
- The weight of the query in a document
- Precision@K
- Recall@K
- Average Precision
- Mean Average Precision
- The throughput of the search engine
- Comparison between Okapi BM25, tf-idf, Wacky formulas
Tools:
- JAVA
- IntelliJ IDEA
- B+tree external library
- Corpus used was from the internet
Find out more
American Sign Language is developed with hand gestures, facial
expression, and motion. In this report, we present the way to detect the alphabets and
digits by using Deep Convolution Neural Network. We also propose to use the pre-defined
Deep Neural Network and compare the accuracy of AlexNet and VGG16 Net. We propose the
implementation of the application to take the video of hand gestures and detect the
alphabet and digits with an accuracy of approximately 98%. The proposed methodology is
scalable and extendable to have multiple classification classes. The proposed
methodology is flexible for taking input with a variety of RGB image sizes. The
classification procedure is followed by pre-processing of images to extract reliable
feature and remove noise if any.
Tools used:
- numpy
- scikit-learn
- sklearn
- Tensorflow-gpu
- Keras
- Opencv-python
- IntelliJ Pycharm
- Jupyter Notebook
Find out more
EBX is an app for Cryptocurrency like Bitcoin, Ripple, etc.
Learning Outcome – 1) Insights of how a Blockchain works. 2) Handling of multiple
crypto-currencies.
Find out more
Sales Kraft is the Sales Team Automation System automatically records all
the stages in a sales process. Sales Kraft includes a contact management system which
tracks all contact that has been made with a given customer the purpose of the contact
and any follow up that may be needed. This ensures that sales efforts are not duplicated
Sales Kraft tracks the sales person and also gives him his tour plan for the day.
Setting targets and objectives based on inputs SALES KRAFT is software that
automatically records all stage in a sales process.
Sales Kraft Includes a contact management system which tracks all contact that has been
made. Sales Kraft Track the Sales Person and also gives him his Tour Plan for the day
Setting targets and objective based on inputs Many salespersons are always on the
go.
• My Task: To develop an app for a salesperson to keep his daily aims, activities,
targets on track.
• Learning Outcome – 1) Insights of how a sales company works. 2) Handling one user on
multiple devices.
Find out more
Syntel Email
• Brief Description – Syntel Email was adapted to become default email
client for Android and iOS devices for management of corporate email in Syntel. It shows
your emails with notifications, similar to Microsoft Outlook and Gmail. It stores and
manages Calendar events, meetings, appointments. It supports auto-sync and manual-sync
feature to sync mail with the support of Syntel MDM Authentication Model.
• Learning Outcome –
1) Worked with the team to design the basic architecture of
the application.
2) Learned security aspects of MNC organization while developing
this application.
Mobile Device Management
• Description - Mobile device management (MDM) is the administrative area
dealing with deploying, securing, monitoring, integrating and managing mobile devices,
such as smartphones, tablets and laptops, in the workplace. The intent of MDM is to
optimize the functionality and security of mobile devices within the enterprise, while
simultaneously protecting the corporate network.
• Learning Outcome - Working with the core functionalities of mobile device.
(Controlling the hardware of the device).
This app is a mobile application for the safety of citizens.
Learning Outcome –1) Interacting with the client face to face. 2) Worked with the
government for people safety.
Find out more
‘Goa Must Go’ is an app developed by a team who love Goa and it is their
intent to provide information & suggestions on the ‘Must Go’ places of Goa.
Features of Goa: Must Go
• Complete information on the attractive tourist attractions with photos and maps
• Phone numbers, email addresses, driving directions of businesses and tourist
attractions
- Beaches
- Hotels
- Restaurants
- Night Clubs
- Shacks
• Emergency services
- Hospitals
- Ambulance Services
- Police
- Airlines
- Railway
- Foreign Consulate Office
• One Touch Search feature to find the best places right next to you, when you are in
Goa.
Find out more