Social Event Detection

Event
detection
is a concept that is crucial to the assurance of public
safety  surrounding  real
-
world
events.  Decision  makers  use  information  from  a
range of terrestrial and online sources to help inform decisions that enable them
to  develop  policies  an
d  react  appropriately  to  events  as  they  unfold.  One  such
source of online information is social media.
Twitter, as a form of social media,
is  a  popular  micro
-
blogging  web  application  serving  hundreds  of  millions  of
users. User
-
generated content can be utilized as a rich source
of information to
identify  real
-
world
events
.
In  this  paper,
we  present  a
novel  d
etection
framework for identifying
such
events,
with a focus on
‘
disruptive
’
events
using
Twitter
data
.
The  approach  is  based  on  five  steps;  data  collection,  pre
-
processing, classification, clustering and summarization. We use a Naïve Bayes
classification
model  and  an  Online  Clustering  method  to  validate  our  model
over  multiple
real
-
world
dat
a
sets
.  To  the  best  of  our  knowledge,  this  study  is
the first effort to identify real
-
world
events in Arabic from social media

In Wireless Sensor Networks (WSN) when an event is detected there is an increase in data traffic that might lead to packets being transmitted through the network close to the packet handling capacity of the WSN. The WSN experiences a decrease in network performance due to packet loss, long delays, and reduction in throughput. In this paper we developed an adaptive congestion control algorithm that monitors network utilization and adjust traffic levels and/or increases network resources to improve throughput and conserve energy. The traffic congestion control protocol DelStatic is developed by introducing backpressure mechanism into NOAH. We analyzed various routing protocols and established that DSR has a higher resource congestion control capability. The proposed protocol, ACCP uses a sink switching algorithm to trigger DelStatic or DSR feedback to a congested node based on its Node Rank. From the simulation results, ACCP protocol does not only improve throughput but also conserves energy which is critical to sensor application survivability on the field. Our Adaptive Congestion control achieved reliability, high throughput and energy efficiency.

В  статье  автор  предлагает  оригинальную  версию  разрешения  проблемы  атомарно-сти социальных событий. Актуальность темы обусловлена тем, что именно недели-мость  позволяет  различить  событие  от  других  социальных  феноменов/процессов.  Событие, с точки зрения автора, должно иметь определенную длительность, которая является атомарной. На первом шаге автор, опираясь на широкий круг источников, включающих взгляды различных теоретиков, рассматривает проблему неделимости социальных событий в сложившейся теории событий. Автор отмечает, что логико-семантические трактовки неделимости событий получили большее распространение, чем утверждение онтологической атомарности. Далее автор подробно останавлива-ется на трактовках атомарности по наблюдению. Анализируя взгляды Д. Дэвидсона, А. Ф. Филиппова и др., автор доказывает, что в случае атомарности по наблюдению: 1) довольно размытыми выглядят критерии этой атомарности; 2) событие оказыва-ется  следствием  не  только  фигуры  наблюдателя,  но  также  его  системы  различения  и мотивов; 3) сложности могут быть связаны с пространственным фактором; 4) само событие смешивается с фактом. Также автор считает не соответствующим современ-ным тенденциям ограничение социальных событий только тем, что доступно взгляду человека.  Далее  автор  показывает  продуктивность  осмысления  события  в  качестве  объекта. В отличие от участников дискуссий, направленных на различение события и  объекта,  автор  использует  концептуализацию  объекта,  предлагаемую  объектно-ориентированной онтологией. События как объекты способны к изменениям, но со-храняют свою неделимость и стабильность, могут быть связаны не только с матери-альной, но и идеальной сферой, обладают системой различений. Предлагаемая схема позволяет утверждать невозможность негативных событий, дает основание не только говорить об атомарности события, но и подчеркивает онтологические основы этой атомарности, а также предлагает возможности мышления о социальном за пределами наблюдаемого.

Geotagged images (e.g., Flickr) could indicate some social events (e.g., festival, parade, protest, sports, etc.) with spatial, temporal and semantic information. Previous researches relied much on tag frequency, thus some images which do not have clearly tag indicating the occurrence of social event would be missing in this case. One potential way to address this problem or enhance the event identification is to make more use of the spatial and temporal information. In this chapter, we take into consideration the underlying spatio-temporal pattern of social events. Particularly, the influence of urban land use and road on the occurrence of event is considered. Specifically, with a spatio-temporal cluster detection method, we firstly detected spatio-temporal clusters composed of geotagged images. Among these detected S-T clusters, we furthermore attempted to identify social events in terms of a classification model. Specifically, land use and road were considered to generate new kinds of spatial characteristics used as dependent variables incorporated into the classification model. In addition to this, user characteristics (i.e., the number of images and the number of users), spatial and temporal range of images, and the heterogeneity of temporal distribution of images were considered as the other dependent variables for the classification model. Consequently, with a binary logistic regression (BLR) method, we estimated the categories (i.e., 'event' or 'non-event' one) of the S-T clusters (cases). Experimental results demonstrated the good performance of the method with a total accuracy of 71 %. With the variable selection process of the BLR method, empirical result also indicates that (1) some characteristics (e.g., the distance to the road and the heterogeneity of temporal distribution of images) do not have considerable influence on the occurrence of 'event'; and (2) compared to the other urban land categories (i.e., residential and recreational land), commercial land has a relatively high influence on the occurrence of 'event'.

Online social networks turn out to be a potential data source to discover worthwhile information from microblogs. Conversely, time-critical exploration of microblog data during catastrophic events such as flood, cyclone, forest fire, and violence carries critical challenges to machine learning techniques. For instance, the microblog from Twitter is utilized to identify event along with its location-specific orientation. In this paper, convolution neural network (CNN) technique is used to identify the retrospective event from the microblog. The existing state of-the-art classification methodologies require substantial volume of labeled data detailed to an unambiguous event during training phase. In addition, it requires feature to attain better outcomes. During the experiments, the n-gram CNN model is trained from the tweets intended for multi-class tweet classification which was related to the specified events in the past without feature engineering. The proposed CNN shows the event detection high accuracy which in turn yields improved performance when compared with other state-of-the-art methods.

Online social networking has become the
predominant activity in the digital world thanks to multimedia
data (mainly photos) sharing (e.g., photos now represent 93% of
the top posts on Facebook). Discovering events where users are
involved using their own posts and those shared by their friends
would be of great importance. In this paper, we address this issue
by providing an original approach able to detect, enrich and also
link user’s events using photos shared within his online social
networks. Using metadata, our approach provides a multidimensional
gathering of similar photos using their temporal,
geographical, and social facets. To validate our approach, we
implemented a prototype called Foto2Events and conducted a set
of experiments on real and generated data. Results show that our
approach works well for various metadata distributions.

Tracklet plots (TPs) describe the motion patterns of a small crowd or a large group of people in a given short time span. This feature can be useful in the context of a Bag-of-Words modelling for the recognition of events or actions that unfold in the scene. This work describes a method where evidence from multiple viewpoints is combined. By obtaining this feature for each of the views, and synchronising the available video streams, a feature-level fusion method by concatenation can be effortlessly applied. The presented system is able to recognise specific events in large groups of people from multiple cameras, and to perform equally well as compared to the best single view available. Furthermore, the dimension of the concatenated feature can be reduced by one order of magnitude without loss of performance.

Large scale social events that involve violence may have dramatic political, economic and social consequences. These events may result in higher crime rates, spreading of infectious diseases, economic crises, and even in migration phenomena (e.g., refugees across borders or internally displaced people). Hence, researchers have started using mobile phone data for developing tools to identify such emergency events in real time. In our paper, we apply a stochastic model, namely a Markov modulated Poisson process, for spatio-temporal detection of hourly and daily behavioral anomalies. We use the call volumes collected from an entire geographic region. Our work is based on the assumption that people tend to make calls when extraordinary events take place. We validate our methodology using a dataset of mobile phone records and events (emergency and non-emergency) from the Republic of Côte d'Ivoire. Our results show that we can successfully capture anomalous calling patterns associated...

We introduce ReDites, a system for realtime event detection, tracking, monitoring and visualisation. It is designed to assist Information Analysts in understanding and exploring complex events as they unfold in the world. Events are automatically detected from the Twitter stream. Then those that are categorised as being security-relevant are tracked, geolocated, summarised and visualised for the end-user. Furthermore, the system tracks changes in emotions over events, signalling possible flashpoints or abatement. We demonstrate the capabilities of ReDites using an extended use case from the September 2013 Westgate shooting incident. Through an evaluation of system latencies, we also show that enriched events are made available for users to explore within seconds of that event occurring.

Events can be characterized by a set of descriptive, collocated keywords extracted documents. Intuitively,
documents describing the same event will contain similar sets of keywords, and the graph for a document
collection will contain clusters individual events. Helping users to understand the event is an acute
problem nowadays as the users are struggling to keep up with tremendous amount of information published
every day in the Internet. The challenging task is to detect the events from online web resources, it is
getting more attentions. The important data source for event detection is a Web search log because the
information it contains reflects users’ activities and interestingness to various real world events. There are
three major issues playing role for event detection from web search logs: effectiveness, efficiency of
detected events. We focus on modeling the content of events by their semantic relations with other events
and generating structured summarization. Event mining is a useful way to understand computer system
behaviors. The focus of recent works on event mining has been shifted to event summarization from
discovering frequent patterns. Event summarization provides a comprehensible explanation of the event
sequence based on certain aspects

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et a ̀ la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Tracklet plots (TPs) describe the motion patterns of a small crowd or a large group of people in a given short time span. This feature can be useful in the context of a Bagof-Words modelling for the recognition of events or actions that unfold in the scene. This work describes a method where evidence from multiple viewpoints is combined. By obtaining this feature for each of the views, and synchronising the available video streams, a feature-level fusion method by concatenation can be effortlessly applied. The presented system is able to recognise specific events in large groups of people from multiple cameras, and to perform equally well as compared to the best single view available. Furthermore, the dimension of the concatenated feature can be reduced by one order of magnitude without loss of performance.

Large scale social events that involve violence may have dramatic political, economic and social consequences. These events may result in higher crime rates, spreading of infectious diseases, economic crises, and even in migration phenomena (e.g., refugees across borders or internally displaced people). Hence, researchers have started using mobile phone data for developing tools to identify such emergency events in real time. In our paper, we apply a stochastic model, namely a Markov modulated Poisson process, for spatio-temporal detection of hourly and daily behavioral anomalies. We use the call volumes collected from an entire geographic region. Our work is based on the assumption that people tend to make calls when extraordinary events take place. We validate our methodology using a dataset of mobile phone records and events (emergency and non-emergency) from the Republic of Côte d'Ivoire. Our results show that we can successfully capture anomalous calling patterns associated with violent events, riots, as well as social non-emergency events such as holidays, sports events. Moreover, call volume changes also show significant temporal and spatial differences depending on the type of an event. Our results provide insights for the long-term goal of developing a real-time event detection system based on mobile phone data.

Twitter's increasing popularity as a source of up-to-date news and information about current events has spawned a body of research on event detection techniques for social media data streams. Although all proposed approaches provide some evidence as to the quality of the detected events, none relate this task-based performance to their run-time performance in terms of processing speed, data throughput, or memory usage. In particular, neither a quantitative nor a comparative evaluation of these aspects has been performed to date. In this article, we study the run-time and task-based performance of several state-of-the-art event detection techniques for Twitter. In order to reproducibly compare run-time performance, our approach is based on a general-purpose data stream management system, whereas task-based performance is automatically assessed based on a series of novel measures.

There are many news articles about events reported on the Web daily, and people are getting more and more used to reading news articles online to know and understand what events happened. For an event, (which may consist of several component events, i.e., episodes), people are often interested in the whole picture of its evolution and development along a time line. This calls for modeling the dependent relationships between component events. Further, people may also be interested in component events which play important roles in the event evolution or development. To satisfy the user needs in finding and understanding the whole picture of an event effectively and efficiently, we formalize in this paper the problem of temporal event search and propose a framework of event relationship analysis for search events based on user queries. We define three kinds of event relationships which are temporal relationship, content dependence relationship, and event reference relationship for identifying to what an extent a component event is dependent on another component event in the evolution of a target event (i.e., query event). Experiments conducted on a real data set show that our method outperforms a number of baseline methods.

Recent advances in integrated electronic devices motivated the use of Wireless Sensor Networks (WSNs) in many applications including target surveillance and tracking. A number of sensor nodes are scattered within a sensitive region to detect the presence of intruders and forward subsequent events to the analysis center(s). Obviously, the sensor deployment should guarantee an optimal event detection rate. This paper proposes a tracking framework based on Voronoi tessellations. Two mobility models are proposed to control the coverage degree according to target presence. The objective is to set a non-uniform coverage within the monitored zone to allow detecting the target(s) by multiple sensor nodes. We show how the proposed algorithm adapts to the situation where multiple targets move in the monitored zone. Moreover, we introduce an algorithm to discover redundant nodes (which do not provide additional information about target position). This algorithm is shown to be effective in reducing the energy consumption using an activity scheduling approach. Simulations are carried out to underline the efficiency of the proposed models.

http://ojs.academypublisher.com/index.php/jnw/article/viewArticle/0407589597