book
Checkout our new book! Hands on AI Trading with Python, QuantConnect, and AWS Learn More arrow

Benzinga

Benzinga News Feed

Introduction

The Benzinga News Feed dataset by Benzinga tracks US Equity news releases. The data covers about 1,250 articles per day across 8,000 Equities, starts in January 2016, and is delivered on a second frequency. This dataset is created by structuring the content produced by Benzinga's editorial team.

This dataset depends on the US Equity Security Master dataset because the US Equity Security Master dataset contains information on splits, dividends, and symbol changes.

For more information about the Benzinga News Feed dataset, including CLI commands and pricing, see the dataset listing.

About the Provider

Benzinga was founded by Jason Raznick in 2010 with goal of connecting the world with news, data, and education that makes the path to financial prosperity easier for everyone, everyday. Benzinga provides access to real-time news for individual investors.

Getting Started

The following snippet demonstrates how to request data from the Benzinga News Feed dataset:

from QuantConnect.DataSource import *

self.aapl = self.add_equity("AAPL", Resolution.DAILY).symbol
self.dataset_symbol = self.add_data(BenzingaNews, self.symbol).symbol
using QuantConnect.DataSource;

_symbol = AddEquity("AAPL", Resolution.Daily).Symbol;
_datasetSymbol = AddData<BenzingaNews>(_symbol).Symbol;

Data Summary

The following table describes the dataset properties:

PropertyValue
Start DateSeptember 2017
Asset Coverage8,000 Equities
Data DensitySparse
ResolutionSecond (1,250 Articles/Day)
TimezoneNew York

Requesting Data

To add Benzinga News Feed data to your algorithm, call the AddDataadd_data method. Save a reference to the dataset Symbol so you can access the data later in your algorithm.

class BenzingaNewsDataAlgorithm(QCAlgorithm):
    def initialize(self) -> None:
        self.set_start_date(2021, 1, 1)
        self.set_end_date(2021, 6, 1)
        self.set_cash(100000)
        
        self.aapl = self.add_equity("AAPL", Resolution.MINUTE).symbol
        self.dataset_symbol = self.add_data(BenzingaNews, self.symbol).symbol
namespace QuantConnect.Algorithm.CSharp.AltData
{
    public class BenzingaNewsDataAlgorithm : QCAlgorithm
    {
        private Symbol _symbol, _datasetSymbol;
        
        public override void Initialize()
        {
            SetStartDate(2021, 1, 1);
            SetEndDate(2021, 6, 1);
            SetCash(100000);
            
            _symbol = AddEquity("AAPL", Resolution.Minute).Symbol;
            _datasetSymbol = AddData<BenzingaNews>(_symbol).Symbol;
        }
    }
}

Accessing Data

To get the current Benzinga News Feed data, index the current Slice with the dataset Symbol. Slice objects deliver unique events to your algorithm as they happen, but the Slice may not contain data for your dataset at every time step. To avoid issues, check if the Slice contains the data you want before you index it.

def on_data(self, slice: Slice) -> None:
    if slice.contains_key(self.dataset_symbol):
        article = slice[self.dataset_symbol]
        self.log(f"{self.dataset_symbol} title at {slice.time}: {article.title}")
public override void OnData(Slice slice)
{
    if (slice.ContainsKey(_datasetSymbol))
    {
        var article = slice[_datasetSymbol];
        Log($"{_datasetSymbol} title at {slice.Time}: {article.Mentions}");
    }
}

To iterate through all of the dataset objects in the current Slice, call the Getget method.

def on_data(self, slice: Slice) -> None:
    for dataset_symbol, article in slice.get(BenzingaNews).items():
        self.log(f"{dataset_symbol} title at {slice.time}: {article.title}")
public override void OnData(Slice slice)
{
    foreach (var kvp in slice.Get<BenzingaNews>())
    {
        var datasetSymbol = kvp.Key;
        var article = kvp.Value;
        Log($"{datasetSymbol} title at {slice.Time}: {article.Title}");
    }
}

Historical Data

To get historical Benzinga News Feed data, call the Historyhistory method with the dataset Symbol. If there is no data in the period you request, the history result is empty.

# DataFrame
history_df = self.history(self.dataset_symbol, 100, Resolution.DAILY)

# Dataset objects
history_bars = self.history[BenzingaNews](self.dataset_symbol, 100, Resolution.DAILY)
var history = History<BenzingaNews>(_datasetSymbol, 100, Resolution.Daily);

For more information about historical data, see History Requests.

Remove Subscriptions

To remove a subscription, call the RemoveSecurityremove_security method.

self.remove_security(self.dataset_symbol)
RemoveSecurity(_datasetSymbol);

If you subscribe to Benzinga News Feed data for assets in a dynamic universe, remove the dataset subscription when the asset leaves your universe. To view a common design pattern, see Track Security Changes.

Example Applications

The Benzinga News Feed enables you to accurately design strategies harnessing real-time news releases. Examples include the following strategies:

  • Creating a dictionary of sentiment scores for various words and assigning a sentiment score to the content of each news release
  • Calculating the sentiment of news releases with Natural Language Processing (NLP)
  • Trading securities when their news releases that Benzinga tags with current buzzwords

Classic Algorithm Example

The following example algorithm parses the Benzinga news articles related to Apple. If the sentiment is positive, the algorithm buys Apple. Otherwise, it holds cash.

from AlgorithmImports import *
from QuantConnect.DataSource import *

class BenzingaNewsDataAlgorithm(QCAlgorithm):

    current_holdings = 0
    target_holdings = 0
    # A custom word-score map for calculating the total sentiment score
    word_scores = {
        'good': 1, 'great': 1, 'best': 1, 'growth': 1,
        'bad': -1, 'terrible': -1, 'worst': -1, 'loss': -1}

    def initialize(self) -> None:
        self.set_start_date(2021, 1, 1)
        self.set_end_date(2021, 6, 1)
        self.set_cash(100000)
        
        # Requesting data to obtain the updated news for sentiment score calculation
        self.aapl = self.add_equity("AAPL", Resolution.MINUTE).symbol
        self.benzinga_symbol = self.add_data(BenzingaNews, self.aapl).symbol
        
        # Historical data
        history = self.history(self.benzinga_symbol, 14, Resolution.DAILY)
        self.debug(f"We got {len(history)} items from our history request")
        
    def on_data(self, slice: Slice) -> None:
        if slice.contains_key(self.benzinga_symbol):
            # Assign a sentiment score to the news article by specific word appearance scoring
            content_words = slice[self.benzinga_symbol].contents.lower()
            score = 0
            for word, word_score in self.word_scores.items():
                score += (content_words.count(word) * word_score)
            self.target_holdings = int(score > 0)
        
        # Ensure we have AAPL data in the current Slice to avoid stale filling
        if not (slice.contains_key(self.aapl) and slice[self.aapl] is not None and not slice[self.aapl].is_fill_forward):
            return
        
        # Buy or sell if the sentiment has changed from our current holdings
        if self.current_holdings != self.target_holdings:
            self.set_holdings(self.aapl, self.target_holdings)
            self.current_holdings = self.target_holdings
using QuantConnect.DataSource;

namespace QuantConnect.Algorithm.CSharp.AltData
{
    public class BenzingaNewsDataAlgorithm : QCAlgorithm
    {
        private Symbol _aapl;
        private Symbol _benzingaSymbol;
        private int _currentHoldings = 0;
        private int _targetHoldings = 0;
        //  A custom word-score map for calculating the total sentiment score
        private Dictionary<string, int> _wordScores = new Dictionary<string, int>(){
            {"good", 1}, {"great", 1}, {"best", 1}, {"growth", 1},
            {"bad", -1}, {"terrible", -1}, {"worst", -1}, {"loss", -1}
        };
        
        public override void Initialize()
        {
            SetStartDate(2021, 1, 1);
            SetEndDate(2021, 6, 1);
            SetCash(100000);
            
            // Requesting data to obtain the updated news for sentiment score calculation
            _aapl = AddEquity("AAPL", Resolution.Minute).Symbol;
            _benzingaSymbol = AddData<BenzingaNews>(_aapl).Symbol;
            
            // Historical data
            var history = History<BenzingaNews>(_benzingaSymbol, 14, Resolution.Daily);
            Debug($"We got {history.Count()} items from our history request");
        }
        
        public override void OnData(Slice slice)
        {
            if (slice.ContainsKey(_benzingaSymbol))
            {
                // Assign a sentiment score to the news article by specific word appearance scoring
                var contentWords = slice[_benzingaSymbol].Contents.ToLower();
                var score = 0;
                foreach (KeyValuePair<string, int> entry in _wordScores)
                {
                    score += (Regex.Matches(contentWords, entry.Key).Count * entry.Value);
                }
                _targetHoldings = Convert.ToInt32(score > 0);
            }
            
            // Ensure we have AAPL data in the current Slice to avoid stale filling
            if (!(slice.ContainsKey(_aapl) && slice[_aapl] != None && !slice[_aapl].IsFillForward))
            {
                return;
            }
            
            // Buy or sell if the sentiment has changed from our current holdings
            if (_currentHoldings != _targetHoldings)
            {
                SetHoldings(_aapl, _targetHoldings);
                _currentHoldings = _targetHoldings;
            }
        }
    }
}

Framework Algorithm Example

The following example algorithm parses the Benzinga news articles related to Apple. If the sentiment is positive, the algorithm buys Apple. Otherwise, it holds cash.

from AlgorithmImports import *
from QuantConnect.DataSource import *

class BenzingaNewsDataAlgorithm(QCAlgorithm):

    def initialize(self) -> None:
        self.set_start_date(2021, 1, 1)
        self.set_end_date(2021, 6, 1)
        self.set_cash(100000)
        
        symbols = [ Symbol.create("AAPL", SecurityType.EQUITY, Market.USA) ]
        self.add_universe_selection(ManualUniverseSelectionModel(symbols))
        
        self.add_alpha(BenzingaNewsAlphaModel())
        
        self.set_portfolio_construction(EqualWeightingPortfolioConstructionModel())
        
        self.add_risk_management(NullRiskManagementModel())
        
        self.set_execution(ImmediateExecutionModel())
        
            
class BenzingaNewsAlphaModel(AlphaModel):
    
    symbol_data_by_symbol = {}
    # A custom word-score map for calculating the total sentiment score
    word_scores = {'good': 1, 'great': 1, 'best': 1, 'growth': 1,
                   'bad': -1, 'terrible': -1, 'worst': -1, 'loss': -1}
    
    def update(self, algorithm: QCAlgorithm, slice: Slice) -> List[Insight]:
        insights = []
        
        for symbol, symbol_data in self.symbol_data_by_symbol.items():
            if slice.contains_key(symbol_data.benzinga_symbol):
                # Assign a sentiment score to the news article by specific word appearance scoring
                content_words = slice[symbol_data.benzinga_symbol].contents.lower()
                score = 0
                for word, word_score in self.word_scores.items():
                    score += (content_words.count(word) * word_score)
                symbol_data.target_direction = InsightDirection.UP if score > 0 else InsightDirection.FLAT
        
            # Ensure we have security data in the current Slice to avoid stale filling
            if not (slice.contains_key(symbol) and slice[symbol] is not None and not slice[symbol].is_fill_forward):
                continue
        
            # Buy or sell if the sentiment has changed from our current holdings
            if symbol_data.current_direction != symbol_data.target_direction:
                symbol_data.current_direction = symbol_data.target_direction
                insights.append(Insight.price(symbol, timedelta(days=14), symbol_data.target_direction))
        
        return insights
        
    def on_securities_changed(self, algorithm: QCAlgorithm, changes: SecurityChanges) -> None:
        for security in changes.added_securities:
            symbol = security.symbol
            self.symbol_data_by_symbol[symbol] = SymbolData(algorithm, symbol)
        
        for security in changes.removed_securities:
            symbol_data = self.symbol_data_by_symbol.pop(security.symbol, None)
            if symbol_data:
                symbol_data.dispose()
            
class SymbolData:
    
    current_direction = InsightDirection.FLAT
    target_direction = InsightDirection.FLAT
    
    def __init__(self, algorithm: QCAlgorithm, symbol: Symbol) -> None:
        self.algorithm = algorithm
        
        # Requesting data to obtain the updated news for sentiment score calculation
        self.benzinga_symbol = algorithm.add_data(BenzingaNews, symbol).symbol
        
        # Historical data
        history = algorithm.history(self.benzinga_symbol, 14, Resolution.DAILY)
        algorithm.debug(f"We got {len(history)} items from our history request")
        
    def dispose(self) -> None:
        # Unsubscribe from Benzinga news feed for this security to release computational resources
        self.algorithm.remove_security(self.benzinga_symbol)
using QuantConnect.DataSource;

namespace QuantConnect.Algorithm.CSharp.AltData
{
    public class BenzingaNewsDataAlgorithm : QCAlgorithm
    {
        public override void Initialize()
        {
            SetStartDate(2021, 1, 1);
            SetEndDate(2021, 6, 1);
            SetCash(100000);
            
            AddUniverseSelection(
                new ManualUniverseSelectionModel(
                    QuantConnect.Symbol.Create("AAPL", SecurityType.Equity, Market.USA)
            ));
            
            AddAlpha(new BenzingaAlphaModel());
            
            SetPortfolioConstruction(new EqualWeightingPortfolioConstructionModel());
            
            AddRiskManagement(new NullRiskManagementModel());
            
            SetExecution(new ImmediateExecutionModel());
        }
    }
    
    public class BenzingaAlphaModel : AlphaModel
    {
        private Dictionary<Symbol, SymbolData> _symbolDataBySymbol = new Dictionary<Symbol, SymbolData>();
        // A custom word-score map for calculating the total sentiment score
        private Dictionary<string, int> _wordScores = new Dictionary<string, int>(){
            {"good", 1}, {"great", 1}, {"best", 1}, {"growth", 1},
            {"bad", -1}, {"terrible", -1}, {"worst", -1}, {"loss", -1}
        };

        public override IEnumerable<Insight> Update(QCAlgorithm algorithm, Slice slice)
        {
            var insights = new List<Insight>();
            
            foreach (var kvp in _symbolDataBySymbol)
            {
                var symbol = kvp.Key;
                var symbolData = kvp.Value;
                
                if (slice.ContainsKey(symbolData.benzingaSymbol))
                {
                    // Assign a sentiment score to the news article by specific word appearance scoring
                    var contentWords = slice[symbolData.benzingaSymbol].Contents.ToLower();
                    var score = 0;
                    foreach (var entry in _wordScores)
                    {
                        score += (Regex.Matches(contentWords, entry.Key).Count * entry.Value);
                    }
                    symbolData.targetDirection = score > 0 ? InsightDirection.Up : InsightDirection.Flat;
                }
                
                
                // Ensure we have security data in the current Slice to avoid stale filling
                if (!(slice.ContainsKey(symbol) && slice[symbol] != None && !slice[symbol].IsFillForward))
                {
                    continue;
                }
                
                // Buy or sell if the sentiment has changed from our current holdings
                if (symbolData.currentDirection != symbolData.targetDirection)
                {
                    insights.Add(Insight.Price(symbol, TimeSpan.FromDays(14), symbolData.targetDirection));
                    symbolData.currentDirection = symbolData.targetDirection;
                }
            }
            return insights;
        }

        public override void OnSecuritiesChanged(QCAlgorithm algorithm, SecurityChanges changes)
        {
            foreach (var security in changes.AddedSecurities)
            {
                var symbol = security.Symbol;
                _symbolDataBySymbol.Add(symbol, new SymbolData(algorithm, symbol));
            }

            foreach (var security in changes.RemovedSecurities)
            {
                var symbol = security.Symbol;
                if (_symbolDataBySymbol.ContainsKey(symbol))
                {
                    _symbolDataBySymbol[symbol].dispose();
                    _symbolDataBySymbol.Remove(symbol);
                }
            }
        }
    }
    
    
    public class SymbolData
    {
        private QCAlgorithm _algorithm;
        public Symbol benzingaSymbol;
        public InsightDirection currentDirection = InsightDirection.Flat;
        public InsightDirection targetDirection = InsightDirection.Flat;
        
        
        public SymbolData(QCAlgorithm algorithm, Symbol symbol)
        {
            _algorithm = algorithm;
            
            // Requesting data to obtain the updated news for sentiment score calculation
            benzingaSymbol = algorithm.AddData<BenzingaNews>(symbol).Symbol;
            
            // Historical data
            var history = algorithm.History<BenzingaNews>(benzingaSymbol, 14, Resolution.Daily);
            algorithm.Debug($"We got {history.Count()} items from our history request");
        }
        
        public void dispose()
        {
            // Unsubscribe from the Benzinga feed for this security to release computational resources
            _algorithm.RemoveSecurity(benzingaSymbol);
        }
    }
}

Data Point Attributes

The Benzinga News Feed dataset provides BenzingaNews objects, which have the following attributes:

You can also see our Videos. You can also get in touch with us via Discord.

Did you find this page helpful?

Contribute to the documentation: