def ef_plot():
output_file = fld_data_analysis_results%GeneralMethods.get_method_id()+'.png'
data = [d for d in FileIO.iterateJsonFromFile(f_hashtag_spatial_metrics, remove_params_dict=True)]
ltuo_hashtag_and_entropy_and_focus = map(itemgetter('hashtag', 'entropy', 'focus'), data)
mf_norm_focus_to_entropies = defaultdict(list)
for _, entropy, (_, focus) in ltuo_hashtag_and_entropy_and_focus:
mf_norm_focus_to_entropies[round(focus, 2)].append(entropy)
plt.figure(num=None, figsize=(6,3))
x_focus, y_entropy = zip(*[(norm_focus, np.mean(entropies))
for norm_focus, entropies in mf_norm_focus_to_entropies.iteritems()
if len(entropies)>0])
plt.subplots_adjust(bottom=0.2, top=0.9, wspace=0, hspace=0)
plt.scatter(x_focus, y_entropy, s=50, lw=0, c='k')
plt.xlim(xmin=-0.1, xmax=1.1)
plt.ylim(ymin=-1, ymax=9)
plt.xlabel('Mean hashtag focus')
plt.ylabel('Mean hashtag entropy')
plt.grid(True)
savefig(output_file)
ltuo_hashtag_and_r_entropy_and_focus =\
sorted(ltuo_hashtag_and_entropy_and_focus, key=itemgetter(1), reverse=True)
ltuo_hashtag_and_r_entropy_and_s_focus = sorted(ltuo_hashtag_and_r_entropy_and_focus, key=itemgetter(2))
hashtags = zip(*ltuo_hashtag_and_r_entropy_and_s_focus)[0]
print list(hashtags[:20])
print list(reversed(hashtags))[:20]
def plot_correlation_ef_plot(condition, id, hashtags, focuses, entropies, peaks):
TIME_UNIT_IN_SECONDS = 10.*60.
mf_norm_focus_to_entropies = defaultdict(list)
mf_norm_focus_to_peaks = defaultdict(list)
for focus, entropy, peak in zip(focuses,entropies, peaks):
if condition(peak):
mf_norm_focus_to_entropies[round(focus, 2)].append(entropy)
mf_norm_focus_to_peaks[round(focus, 2)].append(peak)
x_focus, y_entropy = zip(*[(norm_focus, np.mean(entropies)) for norm_focus, entropies in mf_norm_focus_to_entropies.iteritems() if len(entropies)>5])
_, z_peak = zip(*[(norm_focus, np.mean(peaks)*TIME_UNIT_IN_SECONDS/60) for norm_focus, peaks in mf_norm_focus_to_peaks.iteritems() if len(peaks)>5])
plt.figure(num=None, figsize=(6,3))
plt.subplots_adjust(bottom=0.2, top=0.9, wspace=0, hspace=0)
cm = matplotlib.cm.get_cmap('cool')
sc = plt.scatter(x_focus, y_entropy, c=z_peak, cmap=cm, s=50, lw=0,)
plt.colorbar(sc)
plt.xlim(xmin=-0.1, xmax=1.1)
plt.ylim(ymin=-1, ymax=9)
plt.xlabel('Mean hashtag focus')
plt.ylabel('Mean hashtag entropy')
plt.grid(True)
savefig(output_file_format%id)
ltuo_hashtag_and_entropy_and_focus = zip(hashtags, entropies, focuses)
ltuo_hashtag_and_r_entropy_and_focus = sorted(ltuo_hashtag_and_entropy_and_focus, key=itemgetter(1), reverse=True)
ltuo_hashtag_and_r_entropy_and_s_focus = sorted(ltuo_hashtag_and_r_entropy_and_focus, key=itemgetter(2))
hashtags = zip(*ltuo_hashtag_and_r_entropy_and_s_focus)[0]
print id, list(hashtags)
print id, list(reversed(hashtags))
def temporal_affinity_vs_distance():
output_file = fld_data_analysis_results%GeneralMethods.get_method_id() + '.png'
DataAnalysis._plot_affinities('adoption_lag')
plt.xlabel('Distance (miles)')
plt.ylabel('Hashtag adoption lag (hours)')
# plt.show()
savefig(output_file)
def significant_nei_utm_ids():
output_folder = fld_google_drive_data_analysis%GeneralMethods.get_method_id()+'/%s.png'
for i, data in enumerate(FileIO.iterateJsonFromFile(f_significant_nei_utm_ids, remove_params_dict=True)):
utm_lat_long = UTMConverter.getLatLongUTMIdInLatLongForm(data['utm_id'])
nei_utm_lat_longs = map(
lambda nei_utm_id: UTMConverter.getLatLongUTMIdInLatLongForm(nei_utm_id),
data['nei_utm_ids']
)
if nei_utm_lat_longs:
output_file = output_folder%('%s_%s'%(utm_lat_long))
plotPointsOnWorldMap(nei_utm_lat_longs,
blueMarble=False,
bkcolor='#CFCFCF',
lw = 0,
color = '#EA00FF',
alpha=1.)
_, m = plotPointsOnWorldMap([utm_lat_long],
blueMarble=False,
bkcolor='#CFCFCF',
lw = 0,
color = '#2BFF00',
s = 40,
returnBaseMapObject=True,
alpha=1.)
for nei_utm_lat_long in nei_utm_lat_longs:
m.drawgreatcircle(utm_lat_long[1],
utm_lat_long[0],
nei_utm_lat_long[1],
nei_utm_lat_long[0],
color='#FFA600',
lw=1.5,
alpha=1.0)
print 'Saving %s'%(i+1)
savefig(output_file)
def content_affinity_vs_distance():
output_file = fld_data_analysis_results%GeneralMethods.get_method_id() + '.png'
DataAnalysis._plot_affinities('similarity')
plt.xlabel('Distance (miles)')
plt.ylabel('Hashtags sharing similarity')
# plt.show()
savefig(output_file)
def iid_vs_cumulative_distribution_and_peak_distribution():
TIME_UNIT_IN_SECONDS = 10.*60.
output_file_format = fld_data_analysis_results%GeneralMethods.get_method_id()+'/%s.png'
ltuo_iid_and_interval_stats = [data for data in
FileIO.iterateJsonFromFile(f_iid_spatial_metrics, remove_params_dict=True)]
ltuo_s_iid_and_interval_stats = sorted(ltuo_iid_and_interval_stats, key=itemgetter(0))
ltuo_s_iid_and_tuo_is_peak_and_cumulative_percentage_of_occurrences = [(data[0], (data[1][0], data[1][2])) for data in ltuo_s_iid_and_interval_stats]
total_peaks = sum([data[1][0] for data in ltuo_s_iid_and_tuo_is_peak_and_cumulative_percentage_of_occurrences])+0.0
x_iids = []
y_is_peaks = []
z_cumulative_percentage_of_occurrencess = []
for (iid, (is_peak, cumulative_percentage_of_occurrences)) in ltuo_s_iid_and_tuo_is_peak_and_cumulative_percentage_of_occurrences[:100]:
print (iid, (is_peak, cumulative_percentage_of_occurrences))
x_iids.append((iid+1)*TIME_UNIT_IN_SECONDS/60)
y_is_peaks.append(is_peak/total_peaks)
z_cumulative_percentage_of_occurrencess.append(cumulative_percentage_of_occurrences)
plt.figure(num=None, figsize=(4.3,3))
plt.subplots_adjust(bottom=0.2, top=0.9, wspace=0, hspace=0)
plt.plot(x_iids, y_is_peaks, marker='o', c='k')
plt.ylabel('Distribution of hashtags')
plt.xlabel('Hashtag peak (minutes)')
plt.grid(True)
plt.xlim(xmax=600)
savefig(output_file_format%'peaks')
plt.clf()
plt.figure(num=None, figsize=(6,3))
plt.subplots_adjust(bottom=0.2, top=0.9, wspace=0, hspace=0)
plt.plot(x_iids, z_cumulative_percentage_of_occurrencess, lw=0, marker='o', c='k')
# plt.xlabel('Minutes')
plt.ylabel('CDF of occurrences')
plt.xlabel('Time (Minutes)')
plt.grid(True)
plt.xlim(xmax=600)
savefig(output_file_format%'cdf_occurrences_peak')
def performanceWithSpamDetection(generateData):
experimentData = defaultdict(dict)
ratios = [0.0,0.4,0.9]
marker = dict([(0.0, 's'), (0.4, 'o'), (0.9, 'd')])
# spammerPercentages = [0.2, 0.01, 0.01]
spammerPercentages = [0.015, 0.015, 0.015]
for iteration in range(10):
for spamDetectionRatio, spammerPercentage in zip(ratios, spammerPercentages):
experimentFileName = spamModelFolder+'performanceWithSpamDetection/%s/%0.3f'%(iteration,spamDetectionRatio)
print experimentFileName
if generateData:
model = MixedUsersModel()
conf = {'model': model, 'numberOfTimeSteps': 100, 'addUsersMethod': User.addUsersUsingRatioWithSpamDetection, 'analysisMethods': [(Analysis.measureRankingQuality, 1)], 'ratio': {'normal': 1-spammerPercentage, 'spammer': spammerPercentage},
# 'spammerMessagingProbability': spammerBudget,
'rankingMethods':[RankingModel.latestMessages, RankingModel.latestMessagesSpamFiltered, RankingModel.popularMessages, RankingModel.popularMessagesSpamFiltered],
'spamDetectionRatio': spamDetectionRatio,
'experimentFileName': experimentFileName}
GeneralMethods.runCommand('rm -rf %s'%experimentFileName);run(**conf)
else:
for data in FileIO.iterateJsonFromFile(experimentFileName):
for ranking_id in data['spammmess']:
if data['currentTimeStep'] not in experimentData[spamDetectionRatio]: experimentData[spamDetectionRatio][data['currentTimeStep']]=defaultdict(list)
experimentData[spamDetectionRatio][data['currentTimeStep']][ranking_id]+=data['spammmess'][ranking_id]
if not generateData:
sdr = {}
for spamDetectionRatio in sorted(experimentData.keys()):
dataToPlot = defaultdict(list)
for timeUnit in experimentData[spamDetectionRatio]:
dataToPlot['x'].append(timeUnit)
for ranking_id in experimentData[spamDetectionRatio][timeUnit]: dataToPlot[ranking_id].append(np.mean(experimentData[spamDetectionRatio][timeUnit][ranking_id]))
sdr[spamDetectionRatio]=dataToPlot
for ranking_id in [RankingModel.LATEST_MESSAGES_SPAM_FILTERED, RankingModel.POPULAR_MESSAGES_SPAM_FILTERED]:
# for ranking_id in [RankingModel.LATEST_MESSAGES, RankingModel.POPULAR_MESSAGES]:
for spamDetectionRatio in ratios:
print ranking_id, spamDetectionRatio
dataY = smooth(sdr[spamDetectionRatio][ranking_id],8)[:len(sdr[spamDetectionRatio]['x'])]
dataX, dataY = sdr[spamDetectionRatio]['x'][10:], dataY[10:]
print 'x', [x-10 for x in dataX]
if spamDetectionRatio==0.0:
print ranking_id, dataY
plt.plot([x-10 for x in dataX], dataY, label='%s'%(labels[ranking_id]), lw=1, marker=marker[spamDetectionRatio])
else:
print ranking_id, dataY
plt.plot([x-10 for x in dataX], dataY, label='%s (%d'%(labels[ranking_id].replace('Filtering', 'Detection'),spamDetectionRatio*100)+'%)', lw=1, marker=marker[spamDetectionRatio])
plt.ylim(ymin=0, ymax=1)
plt.xlim(xmin=0, xmax=75)
# plt.title(ranking_id)
plt.legend()
plt.xlabel('Time', fontsize=16, fontweight='bold')
plt.ylabel('Spamness', fontsize=16, fontweight='bold')
# plt.show()
# plt.savefig('performanceWithSpamDetection_%s.png'%ranking_id)
savefig('performanceWithSpamDetection_%s.png'%ranking_id)
plt.clf()
def top_k_locations_on_world_map():
output_file = fld_data_analysis_results%GeneralMethods.get_method_id() + '.png'
ltuo_location_and_occurrence_count = []
for location_object in\
FileIO.iterateJsonFromFile(f_dense_hashtag_distribution_in_locations, remove_params_dict=True):
ltuo_location_and_occurrence_count.append([
location_object['location'],
location_object['occurrences_count']
])
ltuo_lid_and_r_occurrence_count = sorted(ltuo_location_and_occurrence_count, key=itemgetter(1), reverse=True)
# for i, d in enumerate(ltuo_lid_and_r_occurrence_count):
# print i, d
# exit()
lids = zip(*ltuo_lid_and_r_occurrence_count)[0][:200]
points = map(UTMConverter.getLatLongUTMIdInLatLongForm, lids)
plotPointsOnWorldMap(points, blueMarble=False, bkcolor='#CFCFCF', c='m', lw = 0, alpha=1.)
savefig(output_file)
def peak_stats():
TIME_UNIT_IN_SECONDS = 10.*60.
output_file_format = fld_data_analysis_results%GeneralMethods.get_method_id()+'/%s.png'
data = [d for d in FileIO.iterateJsonFromFile(f_hashtag_spatial_metrics, remove_params_dict=True)]
peaks = map(itemgetter('peak_iid'), data)
peaks = filter(lambda i: i<288, peaks)
ltuo_peak_and_count = [(peak, len(list(ito_peaks)))
for peak, ito_peaks in groupby(sorted(peaks))
]
ltuo_s_peak_and_count = sorted(ltuo_peak_and_count, key=itemgetter(0))
current_count = 0.0
total_count = len(peaks)+0.
print total_count
ltuo_peak_and_cdf = []
for peak, count, in ltuo_s_peak_and_count:
current_count+=count
ltuo_peak_and_cdf.append([(peak+1)*TIME_UNIT_IN_SECONDS/(60.), current_count/total_count ])
x_peaks, y_cdf = zip(*ltuo_peak_and_cdf)
plt.figure(num=None, figsize=(4.3,3))
ax=plt.subplot(111)
ax.set_xscale('log')
plt.subplots_adjust(bottom=0.2, top=0.9, left=0.15)
plt.scatter(x_peaks, y_cdf, c='k', s=50, lw=0)
plt.xlabel('Time (minutes)')
plt.ylabel('CDF')
plt.xlim(xmin=5.)
plt.grid(True)
# plt.show()
savefig(output_file_format%'peak_cdf')
plt.clf()
# plt.figure(num=None, figsize=(4.3,3))
ax=plt.subplot(111)
ax.set_xscale('log')
ax.set_yscale('log')
x_peaks, y_counts = zip(*ltuo_s_peak_and_count)
x_peaks = [(peak+1)*TIME_UNIT_IN_SECONDS/(60.) for peak in x_peaks]
y_counts = [count/total_count for count in y_counts]
plt.scatter(x_peaks, y_counts, c='k', s=50, lw=0)
plt.xlabel('Time (minutes)')
plt.ylabel('Distribution of hashtags')
plt.xlim(xmin=5)
plt.ylim(ymax=1., ymin=0.00005)
plt.grid(True)
savefig(output_file_format%'peak_dist')
def plot_global_influencers(ltuo_model_id_and_hashtag_tag):
tuples_of_boundary_and_boundary_label = [
([[-90,-180], [90, 180]], 'World', 'm'),
]
for model_id, hashtag_tag in ltuo_model_id_and_hashtag_tag:
print model_id, hashtag_tag
tuples_of_location_and_color = []
for boundary, boundary_label, boundary_color in tuples_of_boundary_and_boundary_label:
tuo_location_and_influence_scores = Experiments.load_tuo_location_and_boundary_influence_score(model_id, hashtag_tag, boundary)
tuo_location_and_influence_scores = sorted(tuo_location_and_influence_scores, key=itemgetter(1))[:10]
locations = zip(*tuo_location_and_influence_scores)[0]
for location in locations: tuples_of_location_and_color.append([getLocationFromLid(location.replace('_', ' ')), boundary_color])
locations, colors = zip(*tuples_of_location_and_color)
plotPointsOnWorldMap(locations, blueMarble=False, bkcolor='#CFCFCF', c=colors, lw = 0, alpha=1.)
for _, boundary_label, boundary_color in tuples_of_boundary_and_boundary_label: plt.scatter([0], [0], label=boundary_label, c=boundary_color, lw = 0)
# plt.legend(loc=3, ncol=4, mode="expand",)
# plt.show()
savefig(fld_results%(GeneralMethods.get_method_id()) +'%s_%s.png'%(model_id, hashtag_tag))
def location_influence_plots(model_ids, no_of_bins_for_influence_score=100):
for model_id in model_ids:
output_file_format = fld_results%(GeneralMethods.get_method_id()) + '%s_%s.png'
tuo_input_location_and_label_and_marking_locations = [
# [ '40.6000_-73.9500', 'new_york', ['-23.2000_-46.4000', '-22.4750_-42.7750', '51.4750_0.0000', '33.3500_-118.1750', '29.7250_-97.1500','30.4500_-95.7000']],
['29.7250_-97.1500', 'austin', ['-23.2000_-46.4000', '-22.4750_-42.7750', '51.4750_0.0000', '33.3500_-118.1750', '39.1500_-83.3750','30.4500_-95.7000', '40.6000_-73.9500']],
# ['30.4500_-95.7000', 'college_station', ['-23.2000_-46.4000', '-22.4750_-42.7750', '51.4750_0.0000', '33.3500_-118.1750', '29.7250_-97.1500','30.4500_-95.7000', '40.6000_-73.9500']],
]
tuo_location_and_tuo_neighbor_location_and_influence_score = \
Experiments.load_tuo_location_and_tuo_neighbor_location_and_pure_influence_score(model_id)
for input_location, label, marking_locations in tuo_input_location_and_label_and_marking_locations:
for location, tuo_neighbor_location_and_influence_score in \
tuo_location_and_tuo_neighbor_location_and_influence_score:
if input_location==location:
InfluenceAnalysis._plot_scores(tuo_neighbor_location_and_influence_score, marking_locations, no_of_bins_for_influence_score)
plt.xlim(-1,1); plt.ylim(ymin=0.0)
plt.show()
savefig(output_file_format%(label, model_id))
break
def plot_location_plots_with_zones(ltuo_model_id_and_hashtag_tag, no_of_bins_for_influence_score=100):
output_file_format = fld_results+'/%s_%s.png'
for model_id, hashtag_tag in ltuo_model_id_and_hashtag_tag:
no_of_zones, ltuo_location_and_influence_score_and_zone_id = \
Experiments.get_location_with_zone_ids(model_id, hashtag_tag)
locations, influence_scores, zone_ids = zip(*ltuo_location_and_influence_score_and_zone_id)
# print len(locations)
# print [zone_id for _, _, zone_id in sorted(zip(locations, influence_scores, zone_ids), key=itemgetter(1))]
# exit()
# Plot influence plot
ltuo_location_and_global_influence_score = zip(locations, influence_scores)
max_y_tick = InfluenceAnalysis._plot_scores(ltuo_location_and_global_influence_score, [], no_of_bins_for_influence_score, smooth=True)
# Plot zones
ltuo_influence_score_and_zone_id = zip(influence_scores, zone_ids)
ltuo_zone_id_and_influence_scores = [(zone_id, zip(*ito_tuo_influence_score_and_zone_id)[0])
for zone_id, ito_tuo_influence_score_and_zone_id in
groupby(
sorted(ltuo_influence_score_and_zone_id, key=itemgetter(1)),
key=itemgetter(1)
)
]
ltuo_zone_id_and_tuo_min_influence_score_and_max_influence_score = \
[(zone_id, (min(influence_scores), max(influence_scores))) for zone_id, influence_scores in ltuo_zone_id_and_influence_scores]
ltuo_zone_id_and_tuo_box_start_and_box_width = \
[(zone_id, (min_influence_score, abs(min_influence_score-max_influence_score)))
for zone_id, (min_influence_score, max_influence_score) in
ltuo_zone_id_and_tuo_min_influence_score_and_max_influence_score
]
zone_ids, ltuo_box_start_and_box_width = zip(*ltuo_zone_id_and_tuo_box_start_and_box_width)
zone_colors = [GeneralMethods.getRandomColor() for zone_id in zone_ids]
plt.broken_barh(ltuo_box_start_and_box_width , (0, max_y_tick), facecolors=zone_colors, alpha=0.25, lw=0)
# temp_ltuo_box_start_and_box_width = []
# for box_start, box_width in ltuo_box_start_and_box_width:
# if box_width!=0: temp_ltuo_box_start_and_box_width.append((box_start, box_width))
# else: temp_ltuo_box_start_and_box_width.append((box_start, 0.0001))
# zero_size_cluster_ltuo_box_start_and_box_width = []
# for box_start, box_width in ltuo_box_start_and_box_width:
# if box_width==0: zero_size_cluster_ltuo_box_start_and_box_width.append((box_start, 0.0001))
# plt.broken_barh(zero_size_cluster_ltuo_box_start_and_box_width , (0, max_y_tick), facecolors='r', alpha=0.25, lw=0)
# plt.xlim(xmin=-0.0025, xmax=0.0025)
output_file = output_file_format%(GeneralMethods.get_method_id(), model_id, hashtag_tag)
savefig(output_file)
def hashtag_locations_distribution_loglog():
ltuo_no_of_locations_and_count = []
for data in FileIO.iterateJsonFromFile(f_hashtag_and_location_distribution, remove_params_dict=True):
if data[0]=='location' : ltuo_no_of_locations_and_count.append(data[1:])
output_file = fld_data_analysis_results%GeneralMethods.get_method_id() + '.png'
no_of_locations, counts = zip(*ltuo_no_of_locations_and_count)
plt.figure(num=None, figsize=(4.3,3))
plt.subplots_adjust(bottom=0.2, top=0.9, left=0.17)
ax = plt.subplot(111)
ax.set_xscale('log')
ax.set_yscale('log')
plt.scatter(no_of_locations, counts, c='k')
plt.xlabel('No. of locations')
plt.ylabel('No. of hashtags')
plt.xlim(xmin=1/10, )
plt.ylim(ymin=1/10, )
plt.grid(True)
# plt.show()
savefig(output_file)
def global_influence_plots(ltuo_model_id_and_hashtag_tag, no_of_bins_for_influence_score=100):
marking_locations = [
'18.8500_-98.6000',
# '2.9000_101.5000',
'51.4750_0.0000',
'33.3500_-118.1750',
# '-23.2000_-46.4000',
'-22.4750_-42.7750',
'39.1500_-83.3750',
'40.6000_-73.9500',
'29.7250_-97.1500',
'30.4500_-95.7000'
]
for model_id, hashtag_tag in ltuo_model_id_and_hashtag_tag:
output_file = fld_results%(GeneralMethods.get_method_id()) + '%s_%s.png'%(model_id, hashtag_tag)
tuo_location_and_global_influence_score = Experiments.load_tuo_location_and_boundary_influence_score(model_id, hashtag_tag)
InfluenceAnalysis._plot_scores(tuo_location_and_global_influence_score, marking_locations, no_of_bins_for_influence_score, smooth=True)
plt.ylim(ymin=0.0)
# plt.show()
savefig(output_file)
def utm_ids_on_map():
''' Plots utm ids on world map. The color indicates the
log(total_hashtag_count)
'''
output_file = fld_google_drive_data_analysis%GeneralMethods.get_method_id()+'.png'
ltuo_point_and_total_hashtag_count = []
for utm_object in FileIO.iterateJsonFromFile(f_hashtags_by_utm_id, remove_params_dict=True):
point = UTMConverter.getLatLongUTMIdInLatLongForm(utm_object['utm_id'])
total_hashtag_count = log(utm_object['total_hashtag_count'])
ltuo_point_and_total_hashtag_count.append((point, total_hashtag_count))
points, total_hashtag_counts = zip(*sorted(ltuo_point_and_total_hashtag_count, key=itemgetter(1)))
plotPointsOnWorldMap(points,
blueMarble=False,
bkcolor='#CFCFCF',
c=total_hashtag_counts,
cmap=matplotlib.cm.cool,
lw = 0,
alpha=1.)
savefig(output_file)
def plot_graph(locality_measures, id):
mf_apprx_to_count = defaultdict(float)
for measure in locality_measures:
mf_apprx_to_count[round(measure,3)]+=1
total_hashtags = sum(mf_apprx_to_count.values())
current_val = 0.0
x_measure, y_distribution = [], []
for apprx, count in sorted(mf_apprx_to_count.iteritems(), key=itemgetter(0)):
current_val+=count
x_measure.append(apprx)
y_distribution.append(current_val/total_hashtags)
plt.figure(num=None, figsize=(4.3,3))
plt.subplots_adjust(bottom=0.2, top=0.9, left=0.15, wspace=0)
plt.scatter(x_measure, y_distribution, lw=0, marker='o', c='k', s=25)
plt.ylim(ymax=1.2)
if id!='Coverage': plt.xlabel('%s'%id)
else: plt.xlabel('%s (miles)'%id)
plt.ylabel('CDF')
plt.grid(True)
savefig(output_file_format%('cdf_'+id))
def plot_local_influencers(ltuo_model_id_and_hashtag_tag):
tuples_of_boundary_and_boundary_label = [
([[24.527135,-127.792969], [49.61071,-59.765625]], 'USA', GeneralMethods.getRandomColor()),
([[10.107706,-118.660469], [26.40009,-93.699531]], 'Mexico', GeneralMethods.getRandomColor()),
([[-16.6695,88.409841], [30.115057,119.698904]], 'SE-Asia', GeneralMethods.getRandomColor()),
([[-29.565473,-58.191719], [7.327985,-30.418282]], 'Brazil', GeneralMethods.getRandomColor()),
]
for model_id, hashtag_tag in ltuo_model_id_and_hashtag_tag:
print model_id, hashtag_tag
tuples_of_location_and_color = []
for boundary, boundary_label, boundary_color in tuples_of_boundary_and_boundary_label:
tuo_location_and_influence_scores = Experiments.load_tuo_location_and_boundary_influence_score(model_id, hashtag_tag, boundary)
tuo_location_and_influence_scores = sorted(tuo_location_and_influence_scores, key=itemgetter(1))[:10]
locations = zip(*tuo_location_and_influence_scores)[0]
for location in locations: tuples_of_location_and_color.append([getLocationFromLid(location.replace('_', ' ')), boundary_color])
locations, colors = zip(*tuples_of_location_and_color)
plotPointsOnWorldMap(locations, blueMarble=False, bkcolor='#CFCFCF', c=colors, lw = 0, alpha=1.)
for _, boundary_label, boundary_color in tuples_of_boundary_and_boundary_label: plt.scatter([0], [0], label=boundary_label, c=boundary_color, lw = 0)
plt.legend(loc=3, ncol=4, mode="expand",)
# plt.show()
savefig(fld_results%(GeneralMethods.get_method_id()) +'%s_%s.png'%(model_id, hashtag_tag))
def plot_locations_influence_on_world_map(ltuo_model_id_and_hashtag_tag, noOfInfluencers=10, percentage_of_locations=0.15):
input_locations = [
('40.6000_-73.9500', 'new_york'),
('33.3500_-118.1750', 'los_angeles'),
('29.7250_-97.1500', 'austin'),
('30.4500_-95.7000', 'college_station'),
('-22.4750_-42.7750', 'rio'),
('51.4750_0.0000', 'london'),
('-23.2000_-46.4000', 'sao_paulo')
]
for model_id, hashtag_tag in ltuo_model_id_and_hashtag_tag:
tuo_location_and_tuo_neighbor_location_and_locations_influence_score = \
Experiments.load_tuo_location_and_tuo_neighbor_location_and_locations_influence_score(model_id, hashtag_tag, noOfInfluencers=None, influence_type=InfluenceMeasuringModels.TYPE_INCOMING_INFLUENCE)
for input_location, label in input_locations:
for location, tuo_neighbor_location_and_locations_influence_score in \
tuo_location_and_tuo_neighbor_location_and_locations_influence_score:
if input_location==location:
input_location = getLocationFromLid(input_location.replace('_', ' '))
output_file = fld_results%GeneralMethods.get_method_id() + '/%s_%s/%s.png'%(model_id, hashtag_tag, label)
number_of_outgoing_influences = int(len(tuo_neighbor_location_and_locations_influence_score)*percentage_of_locations)
if number_of_outgoing_influences==0: number_of_outgoing_influences=len(tuo_neighbor_location_and_locations_influence_score)
locations = zip(*tuo_neighbor_location_and_locations_influence_score)[0][:number_of_outgoing_influences]
locations = [getLocationFromLid(location.replace('_', ' ')) for location in locations]
# locations = filter(lambda location: isWithinBoundingBox(location, PARTIAL_WORLD_BOUNDARY), locations)
if locations:
_, m = plotPointsOnWorldMap(locations, resolution='c', blueMarble=False, bkcolor='#000000', c='#FF00FF', returnBaseMapObject=True, lw = 0)
# _, m = plotPointsOnWorldMap(locations, resolution='c', blueMarble=False, bkcolor='#CFCFCF', c='#FF00FF', returnBaseMapObject=True, lw = 0)
for location in locations:
# if isWithinBoundingBox(location, PARTIAL_WORLD_BOUNDARY):
m.drawgreatcircle(location[1], location[0], input_location[1], input_location[0], color='#FAA31B', lw=1., alpha=0.5)
# plotPointsOnWorldMap([input_location], blueMarble=False, bkcolor='#CFCFCF', c='#003CFF', s=40, lw = 0)
plotPointsOnWorldMap([input_location], resolution='c', blueMarble=False, bkcolor='#000000', c='#003CFF', s=40, lw = 0)
# plotPointsOnWorldMap([input_location], resolution='c', blueMarble=False, bkcolor='#CFCFCF', c='#003CFF', s=40, lw = 0)
FileIO.createDirectoryForFile(output_file)
print output_file
savefig(output_file)
plt.clf()
else:
GeneralMethods.runCommand('rm -rf %s'%output_file)
break
def fraction_of_occurrences_vs_rank_of_location():
output_file = fld_data_analysis_results%GeneralMethods.get_method_id() + '.png'
ltuo_location_and_occurrence_count = []
for location_object in\
FileIO.iterateJsonFromFile(f_dense_hashtag_distribution_in_locations, remove_params_dict=True):
ltuo_location_and_occurrence_count.append([
location_object['location'],
location_object['occurrences_count']
])
# ltuo_location_and_occurrence_count.sort(key=itemgetter(1))
# for location, occurrence_count in ltuo_location_and_occurrence_count:
# print location, occurrence_count
# exit()
total_occurrences = sum(zip(*ltuo_location_and_occurrence_count)[1]) + 0.0
ltuo_lid_and_r_occurrence_count = sorted(ltuo_location_and_occurrence_count, key=itemgetter(1), reverse=True)
y_fraction_of_occurrences = [r_occurrence_count/total_occurrences for _, r_occurrence_count in ltuo_lid_and_r_occurrence_count]
# total_locations = len(y_fraction_of_occurrences)+0.
# x_percentage_of_locations = [x/total_locations for x in range(1,len(y_fraction_of_occurrences)+1)]
x_percentage_of_locations = range(1,len(y_fraction_of_occurrences)+1)
plt.figure(num=None, figsize=(6,3))
plt.subplots_adjust(bottom=0.2, top=0.9)
plt.semilogy(x_percentage_of_locations, y_fraction_of_occurrences, lw=0, marker='o', c='k')
plt.ylabel('Fraction of occurrences')
plt.xlabel('Locations ordered by their ranks')
plt.grid(True)
a = plt.axes([.55, .5, .3, .3])
# plt.plot(range(10))
plt.semilogy(x_percentage_of_locations, y_fraction_of_occurrences, lw=0, marker='o', c='k')
# plt.title('Probability')
plt.grid(True)
yticks = plt.yticks()
plt.yticks([yticks[0][-1], yticks[0][0]])
# plt.ylim(ymin=0.000001, ymax=0.15)
# plt.ylim(ymin=-0.01, ymax=0.04)
plt.xlim(xmin=-4, xmax=200)
plt.setp(a)
# plt.show()
savefig(output_file)
def example_for_caverlee():
# valid_locations = ['18T_585E_4512N', '18T_587E_4514N']
mf_lid_to_location = dict([
('18T_585E_4512N', 'Times Square'),
('18T_587E_4514N', 'Central Park'),
('18T_584E_4511N', 'Penn Station'),
('18T_585E_4511N', 'Empire State Building'),
])
output_file_format = fld_data_analysis_results%GeneralMethods.get_method_id()+'/%s.png'
subplot_num = 221
# plt.figure(num=None, figsize=(6,3))
for data in FileIO.iterateJsonFromFile(f_example_for_caverlee, remove_params_dict=True):
location = data['location']
if location in mf_lid_to_location:
td = timedelta(hours=-5)
ltuo_occ_time_and_count = data['ltuo_occ_time_and_count']
ltuo_occ_time_and_count.sort(key=itemgetter(0))
occ_times, counts = zip(*ltuo_occ_time_and_count)
occ_times = map(datetime.fromtimestamp, occ_times)
occ_times = map(lambda d: d+td, occ_times)
occ_hours = map(lambda d: d.hour, occ_times)
ltuo_occ_hour_and_count = zip(occ_hours, counts)
ltuo_occ_hour_and_count = [(h, sum(zip(*h_c)[1])) for h, h_c in
GeneralMethods.group_items_by(ltuo_occ_hour_and_count, key=itemgetter(0))]
occ_hours, counts = zip(*ltuo_occ_hour_and_count)
total_counts = sum(counts)+0.0
counts = map(lambda c: c/total_counts, counts)
plt.subplot(subplot_num)
# plt.subplots_adjust(bottom=0.2, top=0.9)
subplot_num+=1
plt.plot(occ_hours, counts, color='#EA00FF', lw=1)
plt.fill_between(occ_hours, counts, color='#EA00FF', alpha=0.25)
# plt.ylabel('% of tweets')
plt.xlabel('Time of day')
plt.xlim(xmax=23)
plt.ylim(ymax=0.09)
plot_anchored_text(mf_lid_to_location[location], loc=2)
plt.grid(True)
# savefig(output_file_format%mf_lid_to_location[location].replace(' ', '_'))
savefig(output_file_format%'ny_locations')
def plot_correlation_between_influence_similarity_and_jaccard_similarity(model_ids):
for model_id in model_ids:
mf_influence_type_to_mf_jaccard_similarity_to_influence_similarities = {}
for line_count, (location, tuo_neighbor_location_and_mf_influence_type_and_similarity) in \
enumerate(FileIO.iterateJsonFromFile(tuo_location_and_tuo_neighbor_location_and_mf_influence_type_and_similarity_file%model_id)):
print line_count
for neighbor_location, mf_influence_type_to_similarity in \
tuo_neighbor_location_and_mf_influence_type_and_similarity:
jaccard_similarity = round(mf_influence_type_to_similarity[JACCARD_SIMILARITY], 1)
for influence_type in \
[InfluenceMeasuringModels.TYPE_OUTGOING_INFLUENCE, InfluenceMeasuringModels.TYPE_INCOMING_INFLUENCE]:
if influence_type not in mf_influence_type_to_mf_jaccard_similarity_to_influence_similarities:
mf_influence_type_to_mf_jaccard_similarity_to_influence_similarities[influence_type] = defaultdict(list)
mf_influence_type_to_mf_jaccard_similarity_to_influence_similarities[influence_type][jaccard_similarity]\
.append(mf_influence_type_to_similarity[influence_type])
subplot_id = 211
for influence_type, mf_jaccard_similarity_to_influence_similarities in \
mf_influence_type_to_mf_jaccard_similarity_to_influence_similarities.iteritems():
plt.subplot(subplot_id)
x_jaccard_similarities, y_influence_similarities = [], []
for jaccard_similarity, influence_similarities in \
sorted(mf_jaccard_similarity_to_influence_similarities.iteritems(), key=itemgetter(0)):
influence_similarities=filter_outliers(influence_similarities)
if len(influence_similarities) > 10:
x_jaccard_similarities.append(jaccard_similarity)
y_influence_similarities.append(np.mean(influence_similarities))
rho, p_value = pearsonr(x_jaccard_similarities, y_influence_similarities)
plt.scatter(x_jaccard_similarities, y_influence_similarities,
c = InfluenceMeasuringModels.INFLUENCE_PROPERTIES[influence_type]['color'],
lw=0, s=40)
plt.plot(x_jaccard_similarities, y_influence_similarities,
c = InfluenceMeasuringModels.INFLUENCE_PROPERTIES[influence_type]['color'], lw=2)
if influence_type==InfluenceMeasuringModels.TYPE_INCOMING_INFLUENCE: plt.ylabel('Influencing locations similarity', fontsize=13)
else: plt.ylabel('Influenced locations similarity', fontsize=13)
subplot_id+=1
plt.xlabel('Jaccard similarity', fontsize=13)
savefig('images/%s.png'%GeneralMethods.get_method_id())
def plot_graph(ltuo_locality_measure_and_occurrences_count, id):
mf_normalized_occurrences_count_to_locality_measures = defaultdict(list)
for locality_measure, occurrences_count in \
ltuo_locality_measure_and_occurrences_count:
normalized_occurrence_count =\
int(occurrences_count/ACCURACY_NO_OF_OCCURRANCES)*ACCURACY_NO_OF_OCCURRANCES+ACCURACY_NO_OF_OCCURRANCES
mf_normalized_occurrences_count_to_locality_measures[normalized_occurrence_count].append(
locality_measure
)
x_occurrance_counts, y_locality_measures = [], []
for k in sorted(mf_normalized_occurrences_count_to_locality_measures):
if len(mf_normalized_occurrences_count_to_locality_measures[k]) > 10:
x_occurrance_counts.append(k), y_locality_measures.append(
np.mean(mf_normalized_occurrences_count_to_locality_measures[k])
)
x_occurrance_counts = [x/1000. for x in x_occurrance_counts]
plt.figure(num=None, figsize=(4.3,3.0))
plt.subplots_adjust(bottom=0.2, top=0.9, left=0.15, wspace=0.)
plt.scatter(x_occurrance_counts, y_locality_measures, lw=0, marker='o', c='k', s=50)
plt.xlabel('Hashtag occurrences in thousands')
plt.ylabel('Mean hashtag %s'%id)
plt.grid(True)
savefig(output_file_format%('locality_vs_occurrences_'+id))
def plot_correlation_between_influence_similarity_and_distance(model_ids, distance_accuracy=500):
def get_larger_lid(lid): return getLatticeLid(getLocationFromLid(lid.replace('_', ' ')), 10)
for model_id in model_ids:
mf_influence_type_to_tuo_distance_and_similarity = defaultdict(list)
for line_count, (location, tuo_neighbor_location_and_mf_influence_type_and_similarity) in \
enumerate(FileIO.iterateJsonFromFile(tuo_location_and_tuo_neighbor_location_and_mf_influence_type_and_similarity_file%model_id)):
print line_count
for neighbor_location, mf_influence_type_to_similarity in \
tuo_neighbor_location_and_mf_influence_type_and_similarity:
distance = getHaversineDistance(getLocationFromLid(location.replace('_', ' ')), getLocationFromLid(neighbor_location.replace('_', ' ')))
distance = int(distance)/distance_accuracy*distance_accuracy + distance_accuracy
for influence_type, similarity in mf_influence_type_to_similarity.iteritems():
mf_influence_type_to_tuo_distance_and_similarity[influence_type].append([distance, similarity])
subpot_id = 211
for influence_type in \
[InfluenceMeasuringModels.TYPE_OUTGOING_INFLUENCE, InfluenceMeasuringModels.TYPE_INCOMING_INFLUENCE]:
tuo_distance_and_similarity = mf_influence_type_to_tuo_distance_and_similarity[influence_type]
tuo_distance_and_similarities = [(distance, zip(*ito_tuo_distance_and_similarity)[1])
for distance, ito_tuo_distance_and_similarity in groupby(
sorted(tuo_distance_and_similarity, key=itemgetter(0)),
key=itemgetter(0)
)
]
plt.subplot(subpot_id)
x_distances, y_similarities = [], []
for distance, similarities in tuo_distance_and_similarities:
# similarities=filter_outliers(similarities)
x_distances.append(distance), y_similarities.append(np.mean(similarities))
# x_distances, y_similarities = splineSmooth(x_distances, y_similarities)
plt.semilogy(x_distances, y_similarities, c = InfluenceMeasuringModels.INFLUENCE_PROPERTIES[influence_type]['color'],
lw=2, marker = InfluenceMeasuringModels.INFLUENCE_PROPERTIES[influence_type]['marker'])
plt.ylabel(InfluenceMeasuringModels.INFLUENCE_PROPERTIES[influence_type]['label'], fontsize=13)
subpot_id+=1
plt.xlabel('Distance (Miles)', fontsize=13)
# plt.show()
savefig('images/%s.png'%(GeneralMethods.get_method_id()))
def performanceWithSpamDetectionVaryingPercentageOfSpammers(generateData):
experimentData = defaultdict(dict)
ratios = [0.0,0.4,0.9]
marker = dict([(0.0, 's'), (0.4, 'o'), (0.9, 'd')])
# spammerPercentages = [0.2, 0.01, 0.01]
# spammerPercentages = [0.015, 0.015, 0.015]
for iteration in range(10):
l1 = [spammerPercentage* 0.001 for spammerPercentage in range(1,51)]
l2 = [spammerPercentage* 0.05 for spammerPercentage in range(1,21)]
l3 = [0.01]+l2
spammer_percentages = l3
for spammerPercentage in spammer_percentages:
for spamDetectionRatio, spammerPercentage in zip(ratios, [spammerPercentage]*3):
experimentFileName = spamModelFolder+'performanceWithSpamDetectionVaryingPercentageOfSpammers/%s/%0.3f/%0.3f'%(iteration,spammerPercentage, spamDetectionRatio)
print experimentFileName
if generateData:
model = MixedUsersModel()
conf = {'model': model, 'numberOfTimeSteps': 10, 'addUsersMethod': User.addUsersUsingRatioWithSpamDetection, 'analysisMethods': [(Analysis.measureRankingQuality, 1)], 'ratio': {'normal': 1-spammerPercentage, 'spammer': spammerPercentage},
# 'spammerMessagingProbability': spammerBudget,
'rankingMethods':[RankingModel.latestMessages, RankingModel.latestMessagesSpamFiltered, RankingModel.popularMessages, RankingModel.popularMessagesSpamFiltered],
'spamDetectionRatio': spamDetectionRatio,
'experimentFileName': experimentFileName}
GeneralMethods.runCommand('rm -rf %s'%experimentFileName);run(**conf)
else:
# for data in FileIO.iterateJsonFromFile(experimentFileName):
# for ranking_id in data['spammmess']:
# if data['currentTimeStep'] not in experimentData[spamDetectionRatio]: experimentData[spamDetectionRatio][data['currentTimeStep']]=defaultdict(list)
# experimentData[spamDetectionRatio][data['currentTimeStep']][ranking_id]+=data['spammmess'][ranking_id]
tempData = defaultdict(list)
for data in FileIO.iterateJsonFromFile(experimentFileName):
for ranking_id in data['spammmess']:
tempData[ranking_id]+=data['spammmess'][ranking_id]
if spammerPercentage not in experimentData[spamDetectionRatio]: experimentData[spamDetectionRatio][spammerPercentage]=defaultdict(list)
for ranking_id in tempData:
experimentData[spamDetectionRatio][spammerPercentage][ranking_id]+=tempData[ranking_id]
if not generateData:
sdr = {}
for spamDetectionRatio in sorted(experimentData.keys()):
dataToPlot = defaultdict(list)
# for spammerPercentage in sorted(experimentData[spamDetectionRatio]):
for spammerPercentage in spammer_percentages:
dataToPlot['x'].append(spammerPercentage)
for ranking_id in experimentData[spamDetectionRatio][spammerPercentage]: dataToPlot[ranking_id].append(np.mean(experimentData[spamDetectionRatio][spammerPercentage][ranking_id]))
sdr[spamDetectionRatio]=dataToPlot
# for ranking_id in [RankingModel.LATEST_MESSAGES_SPAM_FILTERED, RankingModel.POPULAR_MESSAGES_SPAM_FILTERED]:
for ranking_id in [RankingModel.LATEST_MESSAGES, RankingModel.POPULAR_MESSAGES]:
for spamDetectionRatio in ratios:
print ranking_id, spamDetectionRatio
# dataY = smooth(sdr[spamDetectionRatio][ranking_id],8)[:len(sdr[spamDetectionRatio]['x'])]
dataY = sdr[spamDetectionRatio][ranking_id][:len(sdr[spamDetectionRatio]['x'])]
# dataX, dataY = sdr[spamDetectionRatio]['x'][10:], dataY[10:]
dataX, dataY = sdr[spamDetectionRatio]['x'], dataY
# dataX, dataY = splineSmooth(dataX, dataY)
# if spamDetectionRatio==0.0: plt.plot([x-10 for x in dataX], dataY, label='%s'%(labels[ranking_id]), lw=1, marker=marker[spamDetectionRatio])
# else: plt.plot([x-10 for x in dataX], dataY, label='%s (%d'%(labels[ranking_id].replace('Filtering', 'Detection'),spamDetectionRatio*100)+'%)', lw=1, marker=marker[spamDetectionRatio])
if spamDetectionRatio==0.0: plt.plot(dataX, dataY, label='%s'%(labels[ranking_id]), lw=1, marker=marker[spamDetectionRatio])
else: plt.plot(dataX, dataY, label='%s after spam detection (%d'%(labels[ranking_id].replace('Filtering', 'Detection'),spamDetectionRatio*100)+'%)', lw=1, marker=marker[spamDetectionRatio])
# plt.show()
# plt.xlim(xmax=0.05)
# plt.ylim(ymax=0.8)
plt.legend(loc=4)
plt.xlabel('Time', fontsize=16, fontweight='bold')
plt.ylabel('Spamness', fontsize=16, fontweight='bold')
# plt.show()
# plt.savefig('performanceWithSpamDetectionVaryingPercentageOfSpammers_%s.png'%ranking_id)
savefig('/Users/krishnakamath/Dropbox/temp/performanceWithSpamDetectionVaryingPercentageOfSpammers_%s.png'%ranking_id)
# plt.show()
plt.clf()
def norm_iid_vs_locality_measuers():
TIME_UNIT_IN_SECONDS = 10.*60.
output_file_format = fld_data_analysis_results%GeneralMethods.get_method_id()+'/%s.png'
ltuo_normalized_iid_and_tuo_prct_of_occurrences_and_entropy_and_focus_and_coverage = \
[data for data in FileIO.iterateJsonFromFile(f_norm_iid_spatial_metrics, remove_params_dict=True)]
x_normalized_iids, y_entropies, y_focuses, y_distance_from_overall_entropy, y_distance_from_overall_focus, y_coverages = \
zip(*sorted([(data[0]*TIME_UNIT_IN_SECONDS/60, data[1][1], data[1][2], data[1][4], data[1][5], data[1][3])
for data in
ltuo_normalized_iid_and_tuo_prct_of_occurrences_and_entropy_and_focus_and_coverage
])
)
plt.figure(num=None, figsize=(4.3,3))
plt.subplots_adjust(bottom=0.2, top=0.9)
plt.subplot(111)
plt.xlim(xmin=-20, xmax=200)
# plt.ylim(ymin=0.5, ymax=1.0)
plt.plot(x_normalized_iids, y_coverages, lw=1, c='k')
plt.scatter(x_normalized_iids, y_coverages, lw=0, marker='o', s=50, c='k')
plt.ylabel('Interval coverage')
plt.xlabel('Minutes since peak')
plt.grid(True)
savefig(output_file_format%'coverage')
plt.clf()
plt.figure(num=None, figsize=(4.3,3))
plt.subplots_adjust(bottom=0.2, top=0.9)
plt.subplot(111)
plt.xlim(xmin=-20, xmax=120)
plt.ylim(ymin=0.55, ymax=0.70)
plt.plot(x_normalized_iids, y_entropies, lw=1, c='k')
plt.scatter(x_normalized_iids, y_entropies, lw=0, marker='o', s=50, c='k')
plt.ylabel('Interval entropy')
plt.xlabel('Minutes since peak')
plt.grid(True)
savefig(output_file_format%'entropy')
plt.clf()
plt.figure(num=None, figsize=(4.3,3))
plt.subplots_adjust(bottom=0.2, top=0.9)
plt.subplot(111)
plt.xlim(xmin=-20, xmax=400)
# plt.ylim(ymin=1, ymax=3)
plt.plot(x_normalized_iids, y_distance_from_overall_entropy, lw=1, c='k')
plt.scatter(x_normalized_iids, y_distance_from_overall_entropy, marker='o', s=50, c='k')
plt.xlabel('Minutes since peak')
plt.ylabel('Distance from overall entropy')
plt.grid(True)
savefig(output_file_format%'distace_from_overall_entropy')
plt.clf()
plt.figure(num=None, figsize=(4.3,3))
plt.subplots_adjust(bottom=0.2, top=0.9)
plt.subplot(111)
plt.xlim(xmin=-20, xmax=120)
plt.ylim(ymin=0.797, ymax=0.84)
plt.plot(x_normalized_iids, y_focuses, lw=1, c='k')
plt.scatter(x_normalized_iids, y_focuses, lw=1, marker='o', s=50, c='k')
plt.xlabel('Minutes since peak')
plt.ylabel('Interval focus')
plt.grid(True)
savefig(output_file_format%'focus')
plt.clf()
plt.figure(num=None, figsize=(4.3,3))
plt.subplots_adjust(bottom=0.2, top=0.9)
plt.subplot(111)
plt.xlim(xmin=-20, xmax=400)
# plt.ylim(ymin=-0.43, ymax=-0.19)
plt.plot(x_normalized_iids, y_distance_from_overall_focus, lw=1, c='k')
plt.scatter(x_normalized_iids, y_distance_from_overall_focus, marker='o', s=50, c='k')
plt.xlabel('Minutes since peak')
plt.ylabel('Distance from overall focus')
plt.grid(True)
savefig(output_file_format%'distace_from_overall_focus')
def coverage_vs_spatial_properties():
output_file_format = fld_data_analysis_results%GeneralMethods.get_method_id()+'_%s.png'
output_text_file_format = fld_data_analysis_results%GeneralMethods.get_method_id()+'/%s.txt'
data = [d for d in FileIO.iterateJsonFromFile(f_hashtag_spatial_metrics, remove_params_dict=True)]
keys = ['entropy', 'focus', 'spread', 'hashtag', 'num_of_occurrenes', 'focus']
ltuo_entropy_focus_coverage_hashtag_occurrence_count_and_focus_location = map(itemgetter(*keys), data)
ltuo_entropy_focus_coverage_hashtag_occurrence_count_and_focus_location =\
map(
lambda (a,b,c,d,e,f): (a,b[1],c,d,e,f[0]),
ltuo_entropy_focus_coverage_hashtag_occurrence_count_and_focus_location
)
# ltuo_entropy_focus_coverage_hashtag_occurrence_count_and_focus_location = [(data[2], data[3][1], data[4], data[0], data[1], data[3][0]) for data in iterateJsonFromFile(input_file)]
mf_coverage_to_entropies = defaultdict(list)
mf_coverage_to_focuses = defaultdict(list)
mf_coverage_boundary_to_tuo_entropy_and_focus_and_hashtag_and_occurrence_count_and_focus_location =\
defaultdict(list)
total_hashtags = len(ltuo_entropy_focus_coverage_hashtag_occurrence_count_and_focus_location)+0.
for entropy, focus, coverage, hashtag, occurrence_count, focus_location in\
ltuo_entropy_focus_coverage_hashtag_occurrence_count_and_focus_location:
coverage = int(coverage/100)*100+100
mf_coverage_to_entropies[coverage].append(entropy)
mf_coverage_to_focuses[coverage].append(focus)
coverage_boundary = 800
if 800<coverage<1600: coverage_boundary=1600
elif 1600<coverage: coverage_boundary=4000
mf_coverage_boundary_to_tuo_entropy_and_focus_and_hashtag_and_occurrence_count_and_focus_location\
[coverage_boundary].append((entropy, focus, hashtag, occurrence_count, focus_location))
for coverage_boundary, ltuo_entropy_and_focus_and_hashtag_and_occurrence_count_and_focus_location in \
mf_coverage_boundary_to_tuo_entropy_and_focus_and_hashtag_and_occurrence_count_and_focus_location\
.iteritems():
ltuo_entropy_and_focus_and_hashtag_and_s_occurrence_count_and_focus_location = \
sorted(ltuo_entropy_and_focus_and_hashtag_and_occurrence_count_and_focus_location, key=itemgetter(3), reverse=True)
for entropy, focus, hashtag, occurrence_count, focus_location in \
ltuo_entropy_and_focus_and_hashtag_and_s_occurrence_count_and_focus_location:
FileIO.writeToFileAsJson(
[hashtag, occurrence_count, entropy, focus, focus_location],
output_text_file_format%coverage_boundary
)
print coverage_boundary,\
len(ltuo_entropy_and_focus_and_hashtag_and_occurrence_count_and_focus_location)/total_hashtags
print 'median entropy: ',\
np.median(zip(*ltuo_entropy_and_focus_and_hashtag_and_s_occurrence_count_and_focus_location)[0])
print 'median focus: ',\
np.median(zip(*ltuo_entropy_and_focus_and_hashtag_and_s_occurrence_count_and_focus_location)[1])
# print 'var entropy: ', np.var(zip(*ltuo_entropy_and_focus_and_hashtag_and_s_occurrence_count_and_focus_location)[0])
# print 'var focus: ', np.var(zip(*ltuo_entropy_and_focus_and_hashtag_and_s_occurrence_count_and_focus_location)[1])
# print 'range entropy: ', getOutliersRangeUsingIRQ(zip(*ltuo_entropy_and_focus_and_hashtag_and_s_occurrence_count_and_focus_location)[0])
# print 'range focus: ', getOutliersRangeUsingIRQ(zip(*ltuo_entropy_and_focus_and_hashtag_and_s_occurrence_count_and_focus_location)[1])
x_coverages, y_entropies = zip(*[(coverage, np.mean(entropies))
for coverage, entropies in mf_coverage_to_entropies.iteritems()
if len(entropies) > 250])
x_coverages, y_focuses = zip(*[(coverage, np.mean(focuses))
for coverage, focuses in mf_coverage_to_focuses.iteritems()
if len(focuses) > 250])
plt.figure(num=None, figsize=(4.3,3))
ax = plt.subplot(111)
plt.subplots_adjust(bottom=0.2, top=0.9, left=0.15)
plt.scatter(x_coverages, y_entropies, lw=0, marker='o', c='k', s=25)
# plt.ylim(ymax=1.2)
plt.xlabel('Spread (miles)')
plt.ylabel('Entropy')
# ax.set_xscale('log')
plt.grid(True)
savefig(output_file_format%'entropy')
plt.figure(num=None, figsize=(4.3,3))
ax = plt.subplot(111)
plt.subplots_adjust(bottom=0.2, top=0.9, left=0.15)
plt.scatter(x_coverages, y_focuses, lw=0, marker='o', c='k', s=25)
# plt.ylim(ymax=1.2)
plt.xlabel('Spread (miles)')
plt.ylabel('Focus')
# ax.set_xscale('log')
plt.grid(True)
savefig(output_file_format%'focus')