<h1>Terrorist Group Hamas Using BTC To Raise Funds</h1> 0

<h1>Terrorist Group Hamas Using BTC To Raise Funds</h1>

Research from blockchain analysis firm Elliptic shows that terrorist organization Hamas is using bitcoin to raise funds. To hide its activities, the group has developed a website that generates a new wallet after every transaction.

India Banning Bitcoin is Becoming Highly Likely, New Bill Introduced 0

India Banning Bitcoin is Becoming Highly Likely, New Bill Introduced

The Indian inter-ministerial committee has drafted a bill to ban cryptocurrencies like bitcoin.

Subhash Chandra Garg, the Finance Secretary who led the committee, has previously gone on record to warn of the folly of investing in cryptocurrencies. And while the bill is only at the consultation stage, the long term outlook for crypto in India does not look promising.

https://platform.twitter.com/widgets.js

The Indian Authorities Are Not Okay With Bitcoin

The relevant government departments have already received the draft bill titled, the “Banning of Cryptocurrencies and Regulation of Official Digital Currencies Bill 2019.” The government departments will give their feedback to compile the final bill, which the Indian government will consider, following the up-and-coming May elections.

If passed into law, this move would represent the final act in what has been a testing period for the Indian crypto community. Especially so during recent months, which has seen an escalation of anti-crypto sentiment, including a crackdown on exchanges, and the forcible closure of bank accounts held by investors.

In response to the news, prominent crypto advocate and founder of the WazirX exchange, Nischal Shetty remains defiant. In a tweet, he said:

“There’s rumour of India banning crypto. Hope this is fake news. Hundreds of democratic countries allow crypto. As an Indian, I use crypto & I’m not a criminal. Plz listen to the people. RT this, we need to stay united!”

The Indian Government Wants Centralized Control

During late 2016, in a bid to reverse currency counterfeiting, the Indian Prime Minister, Narendra Modi announced a demonetization policy. This set out the withdrawal of Rs 1,000 and Rs 500 currency notes from circulation. As a result, 85% of the country’s money ceased to be legal tender.

The move was widely criticized for being ineffective, for example in not printing sufficient quantities of new notes. But more significantly than that, as a cash-intensive society, the Indian people suffered great hardship during the upheaval process.

Steve Forbes, Chief Editor at Forbes, labeled the move immoral. He said:

“What India has done is commit a massive theft of people’s property without even the pretense of due process–a shocking move for a democratically elected government.”

https://platform.twitter.com/widgets.js

The Impact on Cryptocurrencies

As a result of the demonetization policy, Indians with large sums of cash turned to a new means of holding wealth – Bitcoin. Meaning Modi’s attempt to curb counterfeiting only served to popularize cryptocurrency in India.

Nonetheless, the Indian crypto market is small. Despite the country’s large population, it contributes only 2% of the total cryptocurrency market cap. As such, if the worst were to happen, the impact in the broader crypto market would be slight.

In any case, some say talk of a ban is greatly exaggerated. It can be argued that the bill proposal relates to the implementation of a regulatory framework within India and is something to welcome, in the sense that the Indian authorities have opened dialogue on how to move forward with crypto.

The post India Banning Bitcoin is Becoming Highly Likely, New Bill Introduced appeared first on NewsBTC.

Coinspeaker Partners Blockchain Industry Group: Get Special Discounts Right Now! 0

Coinspeaker Partners Blockchain Industry Group: Get Special Discounts Right Now!

Coinspeaker
Coinspeaker Partners Blockchain Industry Group: Get Special Discounts Right Now!
We are pleased to announce a special partnership between Coinspeaker and the Blockchain Industry Group.
Coinspeaker Partners Blockchain Industry Group: Get Special Discounts Right Now!

Continue reading at Coinspeaker

Presidential Debate Sentiment Analysis with LSTM, OneVsRest, LinearSVC: NLP Step-By-Step Guide 0

Presidential Debate Sentiment Analysis with LSTM, OneVsRest, LinearSVC: NLP Step-By-Step Guide

Cartooned by Alina Z

The data source are tens of thousands of tweets on the first 2016 GOP Presidential Debate in Ohio. What is the sentiment for a given tweet? Is it positive, neutral, or negative? What are the most frequently used words in positive/negative tweets?

You will learn about fundamental Natural Language Processing skills including:

  • Text pre-processing
  • Tokenization
  • Word embedding with TF-IDF
  • Modelling with LSTM, logistic regression, OneVsRest, LinearSVC, etc.
  • Evaluation with F1 score, precision, recall, accuracy

An end-to-end roadmap for NLP projects would be provided at the end of this article.

Load data and take a quick look into the data

import numpy as np 
import pandas as pd
import nltk
nltk.download('stopwords')
from nltk.corpus import stopwords
from sklearn.feature_extraction.text import CountVectorizer
from keras.preprocessing.text import Tokenizer
from keras.preprocessing.sequence import pad_sequences
from keras.models import Sequential
from keras.layers import Dense, Embedding, LSTM, SpatialDropout1D
from keras.utils.np_utils import to_categorical
import re

from sklearn.model_selection import train_test_split, cross_val_score
from sklearn.linear_model import LogisticRegression
from sklearn.svm import LinearSVC
from sklearn.multiclass import OneVsRestClassifier
from sklearn.linear_model import RidgeClassifier

There are 21 columns in the dataset. We only keep the column “text” and “sentiment” here.

Shape of the dataset. We have 13871 rows of record.

There are 3 unique values in the “sentiment” column. Note that the dataset is imbalanced which means the number of records for each category is not equally.

Randomly check a tweet from the dataset.

Split the dataset into random train, validation and test subsets

train_test_split” is a method in scikit-learn that split arrays or matrices into random train and test subsets.

X_train, X_test, y_train, y_test = train_test_split(df['text'], df['sentiment'], test_size=0.33, random_state=42)
X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size=0.2, random_state=42)

Now we have 3 subsets: train, validation, and test. X_train includes the tweets, y_train includes the corresponding sentiment.

Text Pre-processing

Define a function text_prepare for text pre-processing which are handling following tasks:

  • replace symbols in “REPLACE_BY_SPACE_RE” with white space from input text
  • delete symbols in “BAD_SYMBOLS_RE” from input text
  • extend stop words list with ‘rt’, and ‘http’
  • remove stop words from text

https://medium.com/media/b46c04c59b9cabe337cccd38b84f5cd4/href

Process the text in training dataset as follows.

Apply on validation and test dataset.

What are the most common words?

For each word calculate how many times they occur in the train dataset. Sort the dictionary to fetch top 10 common words.

https://medium.com/media/329ef66e40177401fb692d72ff5e5187/href

Word Embedding with TF-IDF

Machine Learning algorithms work with numeric data and we cannot use the provided text data like “@JebBush said he cut FL taxes by $19B”. We need to transform text data to numeric vectors which is called “Word embedding” before feeding them to the models .

vectorization

TF-IDF

The TF-IDF approach (Term Frequency Inverse Document Frequency) extends the bag-of-words framework by taking into account total frequencies of words in the entire dataset of collected tweets. Comparing with bag-of-words, TF-IDF penalizes too frequent words and provides better features space.

  • Use class TfidfVectorizer from scikit-learn
  • Filter out too rare words (occur less than in 5 titles)
  • Filter out too frequent words (occur more than in 90% of the tweets).
  • Use 2-gram along with 1-gram

https://medium.com/media/86a14c2024b1def638a174428d2bdeb8/href

text -> vectors

Finally, we are ready to try out different models.

1st Model: Logistic regression

Use LogisticRegression from sklearn.linear_model

https://medium.com/media/eee1eec8ea17cfa5eb13bfdf0a661bc1/href

LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True, intercept_scaling=1, max_iter=100, multi_class=’warn’, n_jobs=None, penalty=’l2', random_state=None, solver=’warn’, tol=0.0001, verbose=0, warm_start=False)

Cross-validation mean accuracy 67.86%, std 0.38.

2nd Model: LinearSVC

Call LinearSVC from sklearn.svm

https://medium.com/media/c565dc7c18d2186d400b476057e20241/href

Cross-validation mean accuracy 63.76%, std 0.45.

3rd Model: OneVsRest

Call OneVsRestClassifier from sklearn.multiclass

https://medium.com/media/6a129cf2153533a045c46cdcce1da13b/href

Evaluation of the OneVsRestClassifier

https://medium.com/media/18812f3d5dd4187e39d43f64d0ce5b6d/href

Interpretation of evaluation criterion can be found at this document. F1-micro is preferable is preferred because our class is imbalance. Difference between Micro- and macro-averages can be found at this link.

4th Model: LSTM with Keras

Recall that we have imported the keras libraries before.

from sklearn.feature_extraction.text import CountVectorizer
from keras.preprocessing.text import Tokenizer
from keras.preprocessing.sequence import pad_sequences
from keras.models import Sequential
from keras.layers import Dense, Embedding, LSTM, SpatialDropout1D
from keras.utils.np_utils import to_categorical

Save the output of text pre-processing in another pandas data frame “X”.

https://medium.com/media/5f41b96d6cb890245c0a670459c9c542/href

Use Tokenizer from keras

https://medium.com/media/5f41b96d6cb890245c0a670459c9c542/href

Create LSTM model

Pay attention to the activation function and optimizer adam.

https://medium.com/media/f641274b45371a2c8147d39629e3a4a0/href

Encoding the prediction column “sentiment” (Positive, Neutral, Negative)

Train the LSTM model for 20 epochs

https://medium.com/media/9219870416c3b0cd3f75bc2f0684e55a/href

Congrats! You just went through the fundamental technologies in Natural Language Processing including:

  • Text pre-processing
  • Tokenization
  • Word embedding with TF-IDF
  • Modelling with LSTM, logistic regression, OneVsRest, LinearSVC, etc.
  • Evaluation with F1 score, precision, recall, accuracy

The data source used in this project could be found at this link.

The Roadmap of NLP projects can be downloaded at this link.

Next step would be getting your hands dirty by playing with cat. Nope, I mean by coding it up. Good luck.

Sentiment Analysis on Lourdes (my cat)

Presidential Debate Sentiment Analysis with LSTM, OneVsRest, LinearSVC: NLP Step-By-Step Guide was originally published in Hacker Noon on Medium, where people are continuing the conversation by highlighting and responding to this story.

Uber IPO Hype Machine Sputters as Stock Pricing Exposes True Valuation 0

Uber IPO Hype Machine Sputters as Stock Pricing Exposes True Valuation

By CCN: The Uber IPO is one of the most anticipated stock market debuts of the year as the ridesharing company plans to raise $10 billion, but the company’s amended S-1 filing suggests that it won’t be anything more than a damp squib. Uber expects its IPO to be priced in a range of $44 to $50 per share. The ridesharing specialist will be offering up to 180 million shares of its common stock. Those numbers translate into a market cap well below the $100 billion valuation it was reportedly looking for just weeks ago. Uber Slices & Dices IPO

The post Uber IPO Hype Machine Sputters as Stock Pricing Exposes True Valuation appeared first on CCN

<h1>Indian Government May Ban Crypto After All</h1> 0

<h1>Indian Government May Ban Crypto After All</h1>

Several government entities in India have begun talks about drafting a bill to outlaw the purchase, sale, and trade of cryptocurrency in the country and strictly regulate government-approved digital currencies.

Coinme and Coinstar Plan Expansion of Bitcoin ATMs Across 19 U.S. States 0

Coinme and Coinstar Plan Expansion of Bitcoin ATMs Across 19 U.S. States

Coinme has announced that its partnership with Coinstar is paying out major dividends, allowing them to continue expanding their goals to make bitcoin kiosks a commonplace sight in the United States.

The partnership between the two companies began in early 2019, combining Coinme’s bitcoin ATM experience with Coinstar’s nationwide fleet of 20,000 coin-to-cash kiosks.

In a press update on April 24, 2019, Coinme stated that it has experienced “strong national momentum and growth of its mainstream digital currency business, led by a massive increase in the number of kiosks where it offers the ability to buy bitcoin.” Due to the Coinstar partnership, Coinme’s number of bitcoin ATMs has shot up from 70 to more than 2,100 bitcoin-compatible kiosks.

The initial tests of these improved kiosks were carried out in Coinstar terminals set outside small grocery stores across three U.S. states: Washington, California and Texas. Now the plan is to incorporate Coinstar machines at other grocery store chains in a total of 19 states. These new locations include not only metropolitan areas with no previously established Coinme presence, but also a number of urban centers where Coinme has ATMs already operating.

Coinstar CEO Jim Gaherity said that his company is “incredibly pleased with this collaboration with Coinme and [is] eager to continue expanding to new markets in the coming months.”

The success of its expanded ATM operations has resulted in the first quarter of 2019 being “one of the most successful in Coinme’s five-year history, with 38 percent growth in user acquisition, 92 percent gains in transactions volume and 109 percent growth in transactions processed compared to Q1 2018,” according to the company statement.

This article originally appeared on Bitcoin Magazine.

Crypto Scammin’ Church Ministers Steal $2.3 Million as OneCoin Promoters 0

Crypto Scammin’ Church Ministers Steal $2.3 Million as OneCoin Promoters

By CCN.com: For cryptocurrency scammers, no grounds are too holy to scout for potential ‘investors’ as everyone, including churchgoers, is game. Based on an intelligence report from the New Zealand Financial Intelligence Unit (NZFIU), the Central of Bank of Samoa (CBS) has accused two ‘large churches’ in the Asia Pacific (APAC) country of assisting in the proliferation of the OneCoin cryptocurrency scam. Samoan Church Involved in Global Multi-Billion Dollar Cryptocurrency Scam https://t.co/aKlLYPKjKV@SamoanChurch #Global #cryptocurrency #scam — Devdiscourse (@dev_discourse) April 26, 2019 Per the Samoan reserve bank, some of the ministers in the two churches have been acting as promoters of

The post Crypto Scammin’ Church Ministers Steal $2.3 Million as OneCoin Promoters appeared first on CCN

$10 Million in Paxos Stablecoins Printed Overnight Following Tether Allegations 0

$10 Million in Paxos Stablecoins Printed Overnight Following Tether Allegations

More than 10 million Paxos Standard stablecoins were minted after the New York Attorney General’s bombshell allegations about rival issuer Tether.

CheapAir to Accept Ethereum, Dash, Bitcoin Cash and Litecoin 0

CheapAir to Accept Ethereum, Dash, Bitcoin Cash and Litecoin

CheapAir, one of the biggest travel agency that begun accepting bitcoin all the way back in 2013 is now to start accepting ethereum, dash, bitcoin cash and Litecoin. Jeff Klee,…