/*
* Check existence of seed words in a tweet
*/
private int is_exist_seed_words_a_tweet(SDTM_v1_Tweet one_tweet)
{
int target_level_idx = 0;
foreach (KeyValuePair <SDTM_v1_Word, int> one_entry in one_tweet.word_count_table)
{
for (int level_idx = 0; level_idx < SDTM_v1.numLevels; level_idx++)
{
if (this.seed_words_list_level_arr[level_idx].Contains(one_entry.Key.wordidx))
{
target_level_idx = level_idx;
one_entry.Key.seed_word_level = level_idx;
}
}
}
// No seed words in a tweet
return(target_level_idx);
}
/*
* Read conversations from file
*/
private void Read_conversations(string target_file_path)
{
this.convsList = new ArrayList();
int numberofTweets = 0;
int word_count_a_tweet = 0;
string[] line_arr = null;
SDTM_v1_Conversation one_conv = null;
SDTM_v1_Tweet one_tweet = null;
SDTM_v1_Word one_word = null;
Dictionary <SDTM_v1_Word, int> word_count = null;
string one_word_count = null;
string[] one_word_count_arr = null;
int conv_idx = 0;
try
{
using (StreamReader sr = new StreamReader(target_file_path))
{
string line = null;
while ((line = sr.ReadLine()) != null)
{
// Conversation name, userid1_userid2_convid
one_conv = new SDTM_v1_Conversation(conv_idx);
conv_idx++;
line_arr = line.Split('_');
one_conv.set_users(Convert.ToInt32(line_arr[0]), Convert.ToInt32(line_arr[1]));
// number of tweets in the conversation
line = sr.ReadLine();
numberofTweets = Convert.ToInt32(line);
// Each tweet in a conversation
for (int tweet_idx = 0; tweet_idx < numberofTweets; tweet_idx++)
{
// Line format is
// user_id lambda_0 lambda_1 numberofuniquewords BagofWordsFormat
line = sr.ReadLine();
line_arr = line.Split(' ');
one_tweet = new SDTM_v1_Tweet();
one_tweet.set_tweet_id(Convert.ToInt32(line_arr[0]));
one_tweet.set_max_ent_prob(Convert.ToDouble(line_arr[1]), Convert.ToDouble(line_arr[2]));
word_count = new Dictionary <SDTM_v1_Word, int>();
//word_count_a_tweet = Convert.ToInt32(line_arr[3]) + 4; // At the end of line_arr
word_count_a_tweet = line_arr.Length;
for (int word_idx = 4; word_idx < word_count_a_tweet; word_idx++)
{
one_word_count = line_arr[word_idx];
one_word_count_arr = one_word_count.Split(':');
one_word = new SDTM_v1_Word(Convert.ToInt32(one_word_count_arr[0]));
word_count.Add(one_word, Convert.ToInt32(one_word_count_arr[1]));
}
one_tweet.set_word_count(word_count);
// insert tweet to conversation
one_conv.insert_tweet(one_tweet);
}
// insert conversation to list
this.convsList.Add(one_conv);
}
}
}
catch (Exception e)
{
Console.WriteLine(e);
Environment.Exit(1);
}
this.numConvs = this.convsList.Count;
}
public void insert_tweet(SDTM_v1_Tweet one_tweet)
{
this.tweet_list.Add(one_tweet);
}
public void insert_tweet(SDTM_v1_Tweet one_tweet)
{
this.tweet_list.Add(one_tweet);
}
/*
Read conversations from file
*/
private void Read_conversations(string target_file_path)
{
this.convsList = new ArrayList();
int numberofTweets = 0;
int word_count_a_tweet = 0;
string[] line_arr = null;
SDTM_v1_Conversation one_conv = null;
SDTM_v1_Tweet one_tweet = null;
SDTM_v1_Word one_word = null;
Dictionary<SDTM_v1_Word, int> word_count = null;
string one_word_count = null;
string[] one_word_count_arr = null;
int conv_idx = 0;
try
{
using (StreamReader sr = new StreamReader(target_file_path))
{
string line = null;
while ((line = sr.ReadLine()) != null)
{
// Conversation name, userid1_userid2_convid
one_conv = new SDTM_v1_Conversation(conv_idx);
conv_idx++;
line_arr = line.Split('_');
one_conv.set_users(Convert.ToInt32(line_arr[0]), Convert.ToInt32(line_arr[1]));
// number of tweets in the conversation
line = sr.ReadLine();
numberofTweets = Convert.ToInt32(line);
// Each tweet in a conversation
for (int tweet_idx = 0; tweet_idx < numberofTweets; tweet_idx++)
{
// Line format is
// user_id lambda_0 lambda_1 numberofuniquewords BagofWordsFormat
line = sr.ReadLine();
line_arr = line.Split(' ');
one_tweet = new SDTM_v1_Tweet();
one_tweet.set_tweet_id(Convert.ToInt32(line_arr[0]));
one_tweet.set_max_ent_prob(Convert.ToDouble(line_arr[1]), Convert.ToDouble(line_arr[2]));
word_count = new Dictionary<SDTM_v1_Word, int>();
//word_count_a_tweet = Convert.ToInt32(line_arr[3]) + 4; // At the end of line_arr
word_count_a_tweet = line_arr.Length;
for (int word_idx = 4; word_idx < word_count_a_tweet; word_idx++)
{
one_word_count = line_arr[word_idx];
one_word_count_arr = one_word_count.Split(':');
one_word = new SDTM_v1_Word(Convert.ToInt32(one_word_count_arr[0]));
word_count.Add(one_word, Convert.ToInt32(one_word_count_arr[1]));
}
one_tweet.set_word_count(word_count);
// insert tweet to conversation
one_conv.insert_tweet(one_tweet);
}
// insert conversation to list
this.convsList.Add(one_conv);
}
}
}
catch (Exception e)
{
Console.WriteLine(e);
Environment.Exit(1);
}
this.numConvs = this.convsList.Count;
}
/*
Check existence of seed words in a tweet
*/
private int is_exist_seed_words_a_tweet(SDTM_v1_Tweet one_tweet)
{
int target_level_idx = 0;
foreach (KeyValuePair<SDTM_v1_Word, int> one_entry in one_tweet.word_count_table)
{
for (int level_idx = 0; level_idx < SDTM_v1.numLevels; level_idx++)
{
if (this.seed_words_list_level_arr[level_idx].Contains(one_entry.Key.wordidx))
{
target_level_idx = level_idx;
one_entry.Key.seed_word_level = level_idx;
}
}
}
// No seed words in a tweet
return target_level_idx;
}