A series of events in October 2017 led to the initiation of an unprecedented global feminist movement over various social media platforms, where using the hashtag #MeToo (or some variants of it), women across the world publicly shared their untold stories of being sexually harassed. We conducted an anonymous online survey (n=180) and an interview study (n=30) to understand the participation of Bangladeshi women in this movement. Our study concurs that while Bangladeshi women, who are regular users of social media, supported the spirit of this movement; did not participate in it, even though they had many bitter experiences. Our analysis shows that their non-participation was largely influenced by a cultural difference, patriarchy, perceived futility and lack of hope, and a reliance on alternatives. We discuss how our findings of women's use of technology platforms, which is conditioned and limited by male-dominated and conservative Bangladeshi society, relates to the broader issues in feminism that the GROUP community is interested in.

In October 2017, there happened the uprising of an unprecedented online movement on social media by women across the world who started publicly sharing their untold stories of being sexually harassed along with the hashtag #MeToo (or some variants of it). Those stories did not only strike the silence that had long hid the perpetrators, but also allowed women to discharge some of their bottled-up grievances, and revealed many important information surrounding sexual harassment. In this paper, we present our analysis of about one million such tweets collected between October 15 and October 31, 2017 that reveals some interesting patterns and attributes of the people, place, emotions, actions, and reactions related to the tweeted stories. Based on our analysis, we also advance the discussion on the potential role of online social media in breaking the silence of women by factoring in the strengths and limitations of these platforms.

In October 2017, an unprecedented social media movement started where women from all around the world publicly shared their untold stories of being sexually harassed using the hashtag #MeToo (or some variants of it). While this movement raised voices against sexual harassment across the globe, many women in South Asia did not participate in this movement. In this paper, we present our study on non-participation of many Bangladeshi women in this movement through an anonymous online survey (n = 180), and an in-depth interview study (n = 30). Our study shows that many Bangladeshi women, despite being supportive of the movement, did not participate in this movement because of several social, cultural, and infrastructural reasons. We use transnational feminism as a theoretical framework to explain their non-participation. We further discuss how the lessons learned from this study help us better understand the participation, functioning, and impact of social media movements.

Given the increasing popularity of customer service dialogue on Twitter, analysis of conversation data is essential to understanding trends in customer and agent behavior for the purpose of automating customer service interactions. In this work, we develop a novel taxonomy of fine-grained “dialogue acts” frequently observed in customer service, showcasing acts that are more suited to the domain than the more generic existing taxonomies. Using a sequential SVM-HMM model, we model conversation flow, predicting the dialogue act of a given turn in real time, and showcase this using our “PredDial” portal. We characterize differences between customer and agent behavior in Twitter customer service conversations and investigate the effect of testing our system on different customer service industries. Finally, we use a data-driven approach to predict important conversation outcomes: customer satisfaction, customer frustration, and overall problem resolution. We show that the type and location of certain dialogue acts in a conversation have a significant effect on the probability of desirable and undesirable outcomes and present actionable rules based on our findings. We explore the correlations between different dialogue acts and the outcome of the conversations in detail using an actionable-rule discovery task by leveraging a state-of-the-art sequential rule mining algorithm while modeling a set of conversations as a set of sequences. The patterns and rules we derive can be used as guidelines for outcome-driven automated customer service platforms.

Graph classification is a difficult task because finding a good feature representation for graphs is challenging. Existing methods use topological metrics or local subgraphs as features, but the time complexity for finding discriminatory subgraphs or computing some of the crucial topological metrics (such as diameter and shortest path) is high, so existing methods do not scale well when the graphs to be classified are large. Another issue of graph classification is that the number of distinct graphs for each class that are available for training a classification model is generally limited. Such scarcity of graph data resources yields models that have much fewer instances than the model parameters, which leads to poor classification performance. In this work, we propose a novel approach for solving graph classification by using two alternative graph representations: the bag of vertices and the bag of partitions. For the first representation, we use representation learning-based node features and for the second, we use traditional metric-based features. Our experiments with 43 real-life graphs from seven different domains show that the bag representation of a graph improves the performance of graph classification significantly. We have shown 4–75% improvement on the vertex-based and 4–36% improvement on partition-based approach over the existing best methods. Besides, our vertex and partition multi-instance methods are on average 75 and 11 times faster in feature construction time than the current best, respectively.

Being a homeowner is undoubtedly a significant milestone in a person's life. In that pursuit, a prospective buyer spends an enormous amount of time for evaluating potential homes that are available in the market. A study shows that 90% of home buyers rely on the Internet as the primary resource for home related information. However, existing online home search tools i.e. search engines, listing sites, and forums require the user to formulate appropriate search queries to discover the most desired home, which is a complicated task-specifically for the first time home buyers. Another challenge for the home buyers is to filter the search results consisting of hundreds of homes that are generally returned against a search query. With such a process, the perspective home buyer becomes a victim of the well-known “information overload” issue. In this paper, we introduce a new home discovery tool called RAVEN. It uses interactive feedback over a collection of home feature-sets to learn a buyer's interestingness profile. Then it recommends a small list of homes that match with the buyer's interest, thus resolving the “information Overload” problem and eventually decreasing the interval between home search initiation and purchase. Please visit https://youtu.be/e2w3nqM6mnw for a demo video of the system. RAVEN is live at http://bit.ly/2fiZnTo.

Given the increasing popularity of customer service dialogue on Twitter, analysis of conversation data is essential to understand trends in customer and agent behavior for the purpose of automating customer service interactions. In this work, we develop a novel taxonomy of fine-grained "dialogue acts" frequently observed in customer service, showcasing acts that are more suited to the domain than the more generic existing taxonomies. Using a sequential SVM-HMM model, we model conversation flow, predicting the dialogue act of a given turn in real-time. We characterize differences between customer and agent behavior in Twitter customer service conversations, and investigate the effect of testing our system on different customer service industries. Finally, we use a data-driven approach to predict important conversation outcomes: customer satisfaction, customer frustration, and overall problem resolution. We show that the type and location of certain dialogue acts in a conversation have a significant effect on the probability of desirable and undesirable outcomes, and present actionable rules based on our findings. The patterns and rules we derive can be used as guidelines for outcome-driven automated customer service platforms.

MOTIVATION: Moonlighting proteins (MPs) are an important class of proteins that perform more than one independent cellular function. MPs are gaining more attention in recent years as they are found to play important roles in various systems including disease developments. MPs also have a significant impact in computational function prediction and annotation in databases. Currently MPs are not labeled as such in biological databases even in cases where multiple distinct functions are known for the proteins. In this work, we propose a novel method named DextMP, which predicts whether a protein is a MP or not based on its textual features extracted from scientific literature and the UniProt database. RESULTS: DextMP extracts three categories of textual information for a protein: titles, abstracts from literature, and function description in UniProt. Three language models were applied and compared: a state-of-the-art deep unsupervised learning algorithm along with two other language models of different types, Term Frequency-Inverse Document Frequency in the bag-of-words and Latent Dirichlet Allocation in the topic modeling category. Cross-validation results on a dataset of known MPs and non-MPs showed that DextMP successfully predicted MPs with over 91% accuracy with significant improvement over existing MP prediction methods. Lastly, we ran DextMP with the best performing language models and text-based feature combinations on three genomes, human, yeast and Xenopus laevis , and found that about 2.5-35% of the proteomes are potential MPs. AVAILABILITY AND IMPLEMENTATION: Code available at http://kiharalab.org/DextMP . CONTACT: [email protected].

The traditional frequent pattern mining algorithms generate an exponentially large number of patterns of which a substantial proportion are not much significant for many data analysis endeavors. Discovery of a small number of personalized interesting patterns from the large output set according to a particular user's interest is an important as well as challenging task. Existing works on pattern summarization do not solve this problem from the personalization viewpoint. In this work, we propose an interactive pattern discovery framework named PRIIME which identifies a set of interesting patterns for a specific user without requiring any prior input on the interestingness measure of patterns from the user. The proposed framework is generic to support discovery of the interesting set, sequence and graph type patterns. We develop a softmax classification based iterative learning algorithm that uses a limited number of interactive feedback from the user to learn her interestingness profile, and use this profile for pattern recommendation. To handle sequence and graph type patterns PRIIME adopts a neural net (NN) based unsupervised feature construction approach. We also develop a strategy that combines exploration and exploitation to select patterns for feedback. We show experimental results on several real-life datasets to validate the performance of the proposed method. We also compare with the existing methods of interactive pattern discovery to show that our method is substantially superior in performance. To portray the applicability of the framework, we present a case study from the real-estate domain.

The traditional approach of mining frequent patterns generates a very large number of patterns of which a substantial fraction are not much interesting for many data analysis tasks. So selecting a small number of patterns from the large output set such that the selected patterns best align with a particular user's interest is an important task. Existing works on pattern summarization do not help, as these approaches solve interesting pattern discovery from a global perspective which is far from personalization what is needed to meet the pattern discovery demand of a specific user. In this work, we propose an interactive pattern discovery framework, which identifies a set of interesting patterns for a specific user without requiring any prior input on the interestingness measure of patterns from the user. We develop a gradient boosted regression tree based iterative learning algorithm that uses a limited number of interactive feedback from the user to learn her interestingness profile of the patterns, and use this profile for pattern recommendation. We show experimental results on several real-life datasets to validate the performance of the proposed method. We also compare with the existing methods of interactive pattern discovery to show that the performance of the proposed method is substantially superior to the existing methods.

Buying or selling a house is one of the important decisions in a person's life. Online listing websites like "zillow.com", "trulia.com", and "realtor.com" etc. provide significant and effective assistance during the buy/sell process. However, they fail to supply one important information of a house that is, approximately how long will it take for a house to be sold after it first appears in the listing? This information is equally important for both a potential buyer and the seller. With this information the seller will have an understanding of what she can do to expedite the sale, i.e. reduce the asking price, renovate/remodel some home features, etc. On the other hand, a potential buyer will have an idea of the available time for her to react i.e. to place an offer. In this work, we propose a supervised regression (Cox regression) model inspired by survival analysis to predict the sale probability of a house given historical home sale information within an observation time window. We use real-life housing data collected from "trulia.com" to validate the proposed prediction algorithm and show its superior performance over traditional regression methods. We also show how the sale probability of a house is influenced by the values of basic house features, such as price, size, # of bedrooms, # of bathrooms, and school quality.

Frequent subgraph mining (FSM) is an important task for exploratory data analysis on graph data. Over the years, many algorithms have been proposed to solve this task. These algorithms assume that the data structure of the mining task is small enough to fit in the main memory of a computer. However, as the real-world graph data grows, both in size and quantity, such an assumption does not hold any longer. To overcome this, some graph database-centric methods have been proposed in recent years for solving FSM; however, a distributed solution using MapReduce paradigm has not been explored extensively. Since MapReduce is becoming the de-facto paradigm for computation on massive data, an efficient FSM algorithm on this paradigm is of huge demand. In this work, we propose a frequent subgraph mining algorithm called FSM-H which uses an iterative MapReduce based framework. FSM-H is complete as it returns all the frequent subgraphs for a given user-defined support, and it is efficient as it applies all the optimizations that the latest FSM algorithms adopt. Our experiments with real life and large synthetic datasets validate the effectiveness of FSM-H for mining frequent subgraphs from large graph datasets. The source code of FSM-H is available from www.cs.iupui.edu/~alhasan/ software/

Majority of the existing works on network analysis study properties that are related to the global topology of a network. Examples of such properties include diameter, power-law exponent, and spectra of graph Laplacian. Such works enhance our understanding of real-life networks, or enable us to generate synthetic graphs with real-life graph properties. However, many of the existing problems on networks require the study of local topological structures of a network, which did not get the deserved attention in the existing works. In this work, we use graphlet frequency distribution (GFD) as an analysis tool for understanding the variance of local topological structure in a network; we also show that it can help in comparing, and characterizing real-life networks. The main bottleneck to obtain GFD is the excessive computation cost for obtaining the frequency of each of the graphlets in a large network. To overcome this, we propose a simple, yet powerful algorithm, called GRAFT, that obtains the approximate graphlet frequency for all graphlets that have up-to five vertices. Comparing to an exact counting algorithm, our algorithm achieves a speedup factor between 10 and 100 for a negligible counting error, which is, on average, less than 5 percent.

Mining frequent patterns from a hidden dataset is an important task with 43 various real-life applications. In this research, we propose a solution to this problem that is based on Markov Chain Monte Carlo (MCMC) sampling of frequent patterns. Instead of returning all the frequent patterns, the proposed paradigm returns a small set of randomly selected patterns so that the clandestinity of the dataset can be maintained. Our solution also allows interactive sampling, so that the sampled patterns can fulfill the user's requirement effectively. We show experimental results from several real life datasets to validate the capability and usefulness of our solution; in particular, we show examples that by using our proposed solution, an eCommerce marketplace can allow pattern mining on user session data without disclosing the data to the public; such a mining paradigm helps the sellers of the marketplace, which eventually boost the marketplace's own revenue.

Graphlet frequency distribution (GFD) is an analysis tool for understanding the variance of local structure in a graph. Many recent works use GFD for comparing, and characterizing real-life networks. However, the main bottleneck for graph analysis using GFD is the excessive computation cost for obtaining the frequency of each of the graphlets in a large network. To overcome this, we propose a simple, yet powerful algorithm, called GRAFT, that obtains the approximate graphlet frequency for all graphlets that have upto 5 vertices. Comparing to an exact counting algorithm, our algorithm achieves a speedup factor between 10 and 100 for a negligible counting error, which is, on average, less than 5%; For example, exact graphlet counting for ca-AstroPh takes approximately 3 days; but, GRAFT runs for 45 minutes to perform the same task with a counting accuracy of 95.6%.

Graphlet frequency distribution (GFD) has recently become popular for characterizing large networks. However, the computation of GFD for a network requires the exact count of embedded graphlets in that network, which is a computationally expensive task. As a result, it is practically infeasible to compute the GFD for even a moderately large network. In this paper, we propose GUISE, which uses a Markov Chain Monte Carlo (MCMC) sampling method for constructing the approximate GFD of a large network. Our experiments on networks with millions of nodes show that GUISE obtains the GFD within few minutes, whereas the exhaustive counting based approach takes several days.