Tag Archive

Below you'll find a list of all posts that have been tagged as "blog"

Artificial Intelligence Taking over Wall Street trading

One of the biggest reason trading decisions are affected is because of human emotions. Machines and algorithms can make complicated decisions and execute trades at a rate which no human can match and is not influenced by emotions. The parameters these algorithms take into consideration are price variations, macroeconomic data volume changes similar to accounting information of different corporate companies and news articles from various times topredict the nature of a particular stock.Stock prediction can be done using the company's historical data. This historical data can be used to perform either Linear regression or Support Vector Regression depending on the complexity of the system, to discover trends in the stock market. The algorithm can access various real time news papers and journals to retrieve the latest news and information regarding a specific company. This data is then processed and analysed along with the historical data and data derived from the quarterly results and press releases of that company. This helps in predicting a stock price of a specific company.If we need to analyze the whole market, consisting of more than 6000 companies listed in the New York Stock Exchange, we can do that too in the similar manner by navigating through the regulatory filings, social media posts, real time news feed and other finance related metrics also involving elements such as correlations and valuations in order to predict investments which are considered undervalued.AI is already in use by institutional traders and are incorporated in tools used for stock trading. Some of which are completely automated and are used by Hedge Funds. Most of these systems can detect minute changes caused by a number of factors and historical data. As a result thousands of trades are performed on single day.An interesting example:It was noticed that, everytime Anne Hathaway was mentioned in the news, the share price of Berkshire Hathaway increased. This was probably because, there was some algorithm from a trading firm running automatic trades whenever it came across “Hathaway” in the news.This particular example is a false positive and the fact that this system can run automatic trades based on real time news feed is pretty interesting. This technique requires data ingestion, sentiment analysis and entity detection.If the system or algorithm can detect and react to positive news feed faster than anybody else in the market, then one can make the profit that is the leap(or decrease) in price.Citation: http://www.eurekahedge.com/Research/News/1614/Artificial-Intelligence- AI-Hedge-Fund-Index- Strategy-Profile

Aziro Marketing

artificial intelligence blog

How to build an AI app using Tensorflow and Android

Abstract:This article describes a case study on building a mobile app that recognizes objects using machine learning. We have used Tensorflow Lite. Tensorflow Lite machine learning (ML) is an open source library provided by Google. This article mentions a brief on Tensorflow Lite.Object Identification in Mobile app (Creative visualization)Tensorflow Lite:Using Tensorflow, Implement the Machine Learning (ML) or Artificial Intelligence(AI)-powered applications running on mobile phones. ML adds great power to our mobile application. TensorFlow Lite is a lightweight ML library for mobile and embedded devices. TensorFlow works well on large devices and TensorFlow Lite works really well on small devices, as that it’s easier, faster and smaller to work on mobile devices.Machine Learning:Artificial Intelligence is the science for making smart things like building an autonomous driving car or having a computer drawing conclusions based on historical data. It is important to understand that the vision of AI is in ML. ML is a technology where computer can train itself.Neural Network:Neural network is one of the algorithms in Machine learning. One of the use cases of neural networks is, if we have a bunch of images, we can train the neural network to classify which one is the image of a cat or the image of a dog. There are many possible use cases for the combination between ML and mobile applications, starting from image recognition.Machine Learning Model Inside our Mobile Applications:Instead of sending all raw images to the server, we can extract the meaning from the raw data, then send it to the server, so we can get much faster responses from cloud services.This ML model runs inside our mobile application so that mobile application can recognize what kind of object is in each image. So that we can just send the label, such as a cat, dog or human face, to the server. That can reduce the traffic to server. We are going to use Tensorflow Lite in mobile app.TensorFlow Lite Architecture:DEMO: Build an application that is powered by machine learning — Tensorflow Lite in Androidhttps://www.youtube.com/watch?v=olQNKvMbpRgGithub link:Pull below source code, import into Android Studio.https://github.com/Chokkar-G/machinelearningapp.git(Screencast)Tensorflow Lite object detectionThis post contains an example application using TensorFlow Lite for Android App. The app is a simple camera app that classifies images continuously using a pretrained quantized MobileNets model.Workflow :Step 1: Add TensorFlow Lite Android AAR:Android apps need to be written in Java, and core TensorFlow is in C++, a JNI library is provided to interface between the twoThe following lines in the app’s build.gradle file, includes the newest version of the AARbuild.gradle:repositories { maven {url ‘https://google.bintray.com/tensorflow'} } dependencies { // …compile ‘org.tensorflow:tensorflow-lite:+’ }Android Asset Packaging Tool should not compress .lite or .tflite in asset folder, so add following block.android { aaptOptions {noCompress “tflite”noCompress “lite”} }Step 2: Add pretrained model files to the projecta. Download the pretrained quantized Mobilenet TensorFlow Lite model from herehttps://storage.googleapis.com/download.tensorflow.org/models/tflite/mobilenet_v1_224_android_quant_2017_11_08.zipb. unzip and copy mobilenet_quant_v1_224.tflite and label.txt to the assets directory: src/main/assets(Screencast) Placing model file in assets folderStep 3: Load TensorFlow Lite Model:The Interpreter.java class drives model inference with TensorFlow Lite.tflite = new Interpreter(loadModelFile(activity));Step 4: Run the app in device.Conclusion:Detection of objects like a human eye has not been achieved with high accuracy using cameras, i.e., cameras cannot replace a human eye. Detection refers to identification of an object or a person by training a model by itself. However, we do have great potential in MI. This was just a simple demonstration for MI. We could create a robot that changes its behavior and its way of talking according to who’s in front of it (a child/ an adult). We can use deep learning algorithm to identify skin cancer, or detect defective pieces and automatically stop a production line as soon as possible.References:https://www.tensorflow.org/mobile/tflite/https://www.tensorflow.org/mobile/

Aziro Marketing

artificial intelligence blog

How to build a student-centered Feedback system using AI

With the advent of machine learning and Artificial Intelligence every industry be it Healthcare, Education, or Finance, etc. has been disrupted. Technology can be seen as both, Yin and Yang, it can be beneficent and can be malignant and machine learning gives us enormous power to solve complex problems that cannot be possible with writing rule for every specific case.Machine learning has been disrupting Education Technology and has been prolific across most of the EdTech startup. Be it provision of recommendation to students about the courses or assisting students with feedback or summarizing the content so reader can comprehend the useful part of the content rather than getting into nitty gritty details of every text.We thought of building an application in a similar domain, assisting students with feedback and self-evaluation. In the age of information and technology where data is available in abundance, it’s difficult as well as imperative for a student to learn and understand the concept behind the content; regular assessment and keeping the user progress can be a better way to improvise the learning.Let’s start our Journey – Building a student-centered Feedback systemThere are couple of entities a person needs to keep in mind while building Text processing and Natural language system. A complex Natural language system contains a substantial number of algorithm depending on the use case, be it a key phrase extraction using Tf-idf , Text-rank for Text summarization, Bayes Theorem for Sentiment analysis, POS tagging, NER extraction using Naïve Bayes etc. these are some of the simple algorithms which can be useful while building a simple text processing engine. With the advent of deep learning, it is feasible to replenish the area of Natural language understanding which opens a broader scope of understanding the emotion behind the text the user conveyed, such as sarcasm, humor, disgust, excitement etc.This technology can help us build a complex natural language system and general-purpose AI.The components we need for our purpose are:1. Crawler service that crawls each section of the document, create a token of each term, and create a snippet or tag and save it in the database.2. A Reverse indexer that maps document-id to the term found within the content.3. To measure the similarity of the answer with the actual answer while taking the assessment we’ll be using consine similarity matrix. For complex systems you can use deep-learning RNN to measure the similarity.4. We will be using Watson API to understand the confidence of the students while answering the question.The Query EngineThe user answers to the question asked by the system through Quey API, which will be parsed and decomposed into tokens of N vector. Here we split the answer into token and match each with the actual answer token of M vector and find the cosine similarity between both the answers. Consine similarity doesn’t take into account the magnitude between both the answers but the angle between them.If the given answer is above the given threshold we assume that the answer is correct and try to find the confidence/emotion behind the answer using Watson cognitive API.If the answer is incorrect we will decompose the actual answer which is mapped to the question asked to the student into tokens of N vectors and find the document/ section id of the question from which answers are prepared from. To find the section id we will use Parsed Query and Standard Query where we take the intersection of matched token and use Text rank algorithm to rank the matching document/section.Intricacy of NLP system:There are couple of intricacies while building a Text processing system. Scaling a system, avoiding stops words, navigating to correct section of the document, technique and heuristics of parsed queries to recommend actual section id and finally the confidence of the student on answering the questions. We’ll not build the entire system from scratch but we’ll use some of existing library available.Some of the important libraries and API’s we have used are:1. Tensor flow2. Scikit learn3. NLTK to text processing4. Whoosh / Elastic search for reverse indexer5. Watson cognitive APIThis is a general purpose solution for building a feedback system. It can be used in organization to get feedback from employees and it can also be used to get feedback from the customer and get the emotion out of it and avoid the need of manual evaluation of each feedback response.Here I come to an end of the brief overview of building a student feedback system, we have already gone through few of the algorithm necessary to build an NLP system. I suggest readers to search some of the material related to Natural Language Processing to get a better understanding of the subject.

Aziro Marketing

artificial intelligence blog

Using Predictive Artificial Intelligence for the future of Healthcare

Artificial Intelligence (A.I) is used in all the sectors for improvement and better outcome. For example NASA used Google AI to find new planets in the galaxy. This has become a news for the world. Would like to give you a brief introduction on AI on healthcare. Using AI in healthcare will reduce the complications which come into existence when a person goes through an surgery and also can give him best possible information depending or diet to not have a surgery. We could balance the ethics and efficiency of the healthcare industry. By using AI, points which are overlooked or missed in urgency can be taken care at the primary level than at a critical level. Since I came across an health emergency in my family, so thought how I can use AI to contribute in healthcare. I determined the best possible match of Kidney Transplantation, started reading the information present around me to find out the information related to these organs. In my findings I found out that each have many complications can also lead to death of an individual. Transplant option would be thought of when the kidneys stop functioning entirely. This condition is called end-stage renal disease (ESRD) or end-stage kidney disease (ESKD). If you reach this point, your doctor is likely to recommend dialysis. In addition to the dialysis, your doctor will inform you whether you are a good candidate for kidney transplant. You will need to be healthy enough to have major surgery and go through a strict, lifelong medication after surgery to be a good candidate for a transplant. You must also be willing and able to follow all instructions from your doctor and take your medications regularly. If your doctor thinks you’re good candidate for a transplant, and you’re interested in the procedure, you’ll need to be evaluated at a transplant center. This examination usually involves several visits to the hospital to assess your physical, psychological and familial condition. The doctors will run tests on your blood and urine and give you a complete information to ensure you’re healthy enough for surgery. The Matching Process At the time of evaluation for transplant, blood tests would be conducted to determine your blood type (A, B, AB, or O) and your human leukocyte antigen (HLA). HLA is a group of antigens which is located on the surface of your white blood cells, they are responsible for your body’s immune response. If donor and your HLA type matches, then your body will not reject the kidney. Each person has six antigens, three from each biological parent. The more antigens matches you have with the donor, the greater the chance of a successful transplant. Once a best possible donor has been identified, another test has to be take up to ensure sure that your antibodies won’t attack the donor’s organ. This is done by mixing a small amount of your blood with the donor’s blood. The transplant can’t be taken up if your blood forms any antibodies in response to the donor’s blood. If the blood shows no antibody reaction which is called “negative crossmatch”, i.e. transplant can proceed. Algorithm Process During this process, we take the old data related to this and train our engine. So how does the engine work? Engine works but connecting all the hospital present around that town/city/country, it will recognise the best possible chances of the success. We get to know this from the data present with us from the old cases. Our engine reads the information and gets best possible procedure to go ahead. For a simple example, a person in a certain city will be in need to transplantation and he does get suitable match. There is one more person in another city with the same problem. The donor who is ready for that person matches with this person1 and donor of person1 matches with person2. We can plan the best possible match and provide the information to concerned person such that a best possible help will be provided to patients who are in the hospital. The algo helps doctors to get to know the complication which many come to a person while undergoing the procedure such that a solution can be ready while the process is going on. This will reduced the risk of the patients who undergo major operations. So we are looking forward to work on this and make it a best scenario for showing the capabilities of the AI engine which will increase the life expectancy of a human being.

Aziro Marketing

artificial intelligence blog

EXPLORE ALL TAGS

Ansible Test Automation

application containerization

applications

Application Security

application testing

artificial intelligence

asynchronous replication

business intelligence

Cloud Cost Optimization

cloud devops

Cloud Infrastructure

Cloud Interoperability

Cloud Native Solution

Cloud Storage Security

Codeless Automation

Cognitive analytics

Configuration Management

container world conference

continuous-delivery

continuous deployment

continuous integration

data backup and recovery

Descriptive analytics

Descriptive analytics tools

Digital Transformation

dockercon 2019 san francisco

end-to-end-test-automation

High Performance Computing

icinga for monitoring

Image Recognition 2024

kubernetesday bangalore

Low-Code No-Code Platforms

mobile-application-testing

mobile-automation-testing

Predictive analytics tools

prescriptive analysis

Rapid Application Development

raspberry pi

RDMA

real time analytics

realtime analytics platforms

Real-time data analytics

Recovery

Recovery as a service

rsa 2019 san francisco

Selenium Test Automation

selenium testng

serverless

Serverless Computing

Site Reliability Engineering

software defined storage

software-testing

software testing trends

software testing trends 2019

storage virtualization

support

Synchronous Replication

testing automation tools

thought leadership articles

trends

tutorials

ui automation testing

ui testing

ui testing automation

vCenter Operations Manager

vmworld 2019 san francisco

VMworld 2019 US

vROM

Web Automation Testing

web test automation

WFH

Real People, Real Replies.
No Bots, No Black Holes.

Big things at Aziro often start small - a message, an idea, a quick hello. A real human reads every enquiry, and a simple conversation can turn into a real opportunity.
Start yours with us.

Talk to us

+1 844 415 0777

Drop us a line at

info@aziro.com

Tag Archive

Artificial Intelligence Taking over Wall Street trading

How to build an AI app using Tensorflow and Android

How to build a student-centered Feedback system using AI

Using Predictive Artificial Intelligence for the future of Healthcare

EXPLORE ALL TAGS

Real People, Real Replies.No Bots, No Black Holes.

Got a Tech Challenge? Let’s Talk

Real People, Real Replies.
No Bots, No Black Holes.