public static async Task <List <TweetSentiment> > AnalyzeTweetSentiment(List <TweetRaw> tweets)
{
var tweetsSentiment = new List <TweetSentiment>();
int page = 0;
int itemPerPage = 50;
var temp = new List <TweetRaw>();
temp.AddRange(tweets.Skip(itemPerPage * page++).Take(itemPerPage));
while (temp.Count > 0)
{
var textBatchInput = new TextAnalyticsBatchInput();
foreach (var t in temp)
{
var textAnalysis = new TextAnalyticsInput
{
Id = t.Id.ToString(),
Text = t.Text
};
textBatchInput.Documents.Add(textAnalysis);
}
var sentimentResponse = await AzureSentiment.SentimentV3PreviewPredictAsync(textBatchInput);
Console.WriteLine($"Tweets analyzed:{itemPerPage + itemPerPage * (page - 1)}");
foreach (var document in sentimentResponse.Documents)
{
var tweetSentiment = new TweetSentiment
{
TweetRawId = long.Parse(document.Id),
PositiveScore = document.DocumentScores.Positive,
NeutralScore = document.DocumentScores.Neutral,
NegativeScore = document.DocumentScores.Negative,
Sentiment = Enum.Parse <Degree.Models.DocumentSentimentLabel>(document.Sentiment.ToString()),
};
if (document.Sentences != null && document.Sentences.Count() > 0)
{
foreach (var sentence in document.Sentences)
{
var s = new TweetSentenceSentiment()
{
TweetSentimentId = long.Parse(document.Id),
Length = sentence.Length,
PositiveScore = sentence.SentenceScores.Positive,
NeutralScore = sentence.SentenceScores.Neutral,
NegativeScore = sentence.SentenceScores.Negative,
Offset = sentence.Offset,
Sentiment = Enum.Parse <Degree.Models.SentenceSentimentLabel>(sentence.Sentiment.ToString()),
Warnings = sentence.Warnings
};
tweetSentiment.Sentences.Add(s);
}
}
tweetsSentiment.Add(tweetSentiment);
}
temp.Clear();
temp.AddRange(tweets.Skip(itemPerPage * page++).Take(itemPerPage));
}
return(tweetsSentiment);
}
public static async Task <List <TweetEntityRecognized> > AnalyzeTweetEntity(List <TweetRaw> tweets)
{
var tweetsEntity = new List <TweetEntityRecognized>();
int page = 0;
int itemPerPage = 50;
var temp = new List <TweetRaw>();
temp.AddRange(tweets.Skip(itemPerPage * page++).Take(itemPerPage));
while (temp.Count > 0)
{
var textBatchInput = new TextAnalyticsBatchInput();
foreach (var t in temp)
{
var textAnalysis = new TextAnalyticsInput
{
Id = t.Id.ToString(),
Text = t.Text
};
textBatchInput.Documents.Add(textAnalysis);
}
var entityRecognitionResponse = await AzureEntityRecognition.EntityRecognitionV3PreviewPredictAsync(textBatchInput);
Console.WriteLine($"Tweets analyzed:{itemPerPage + itemPerPage * (page - 1)}");
foreach (var document in entityRecognitionResponse.Documents)
{
foreach (var entity in document.Entities)
{
var tweetEntity = new TweetEntityRecognized
{
EntityName = entity.Text,
EntityType = entity.Type,
Length = entity.Length,
Offset = entity.Offset,
Score = entity.Score,
TweetRawId = long.Parse(document.Id)
};
tweetsEntity.Add(tweetEntity);
}
}
temp.Clear();
temp.AddRange(tweets.Skip(itemPerPage * page++).Take(itemPerPage));
}
return(tweetsEntity);
}