Complex Semantic Relationships Research Articles

Listen as you wish: Fusion of audio and text for cross-modal event detection in smart cities

In the era of smart cities, the advent of the Internet of Things technology has catalyzed the proliferation of multimodal sensor data, presenting new challenges in cross-modal event detection, particularly in audio event detection via textual queries. This paper focuses on the novel task of text-to-audio grounding (TAG), aiming to precisely localize sound segments that correspond to events described in textual queries within an untrimmed audio. This challenging new task requires multi-modal (acoustic and linguistic) information fusion as well as the reasoning for the cross-modal semantic matching between the given audio and textual query. Unlike conventional methods that often overlook the nuanced interactions between and within modalities, we introduce the Cross-modal Graph Interaction (CGI) model. This innovative approach leverages a language graph to model complex semantic relationships between query words, enhancing the understanding of textual queries. Additionally, a cross-modal attention mechanism generates snippet-specific query representations, facilitating fine-grained semantic matching between audio segments and textual descriptions. A cross-gating module further refines this process by emphasizing relevant features across modalities and suppressing irrelevant information, optimizing multimodal information fusion. Our comprehensive evaluation on the Audiogrounding benchmark dataset not only demonstrates the CGI model’s superior performance over existing methods, but also underscores the significance of sophisticated multimodal interaction in improving the efficacy of TAG in smart cities.

Употребата на възвратните притежателни местоимения свой, си в устната форма на българския книжовен език

Current normative problems related to the use of the reflexive possessive pronouns svoy, si in the oral form of the Bulgarian literary language are considered. Attention is drawn to non-trivial cases of the use of reflexive possessives instead of possessive pro-nouns. Attention is paid to the competition between the two types of pronouns in certain types of syntactic constructions and the reasons for errors and hesitations are analyzed. Cases that present more complex semantic relations, encoded in the sentence as a linguistic unit, at the level of which the rule for the use of the reflexive possessive pronoun is manifested: sentences with an extended noun phrase, with nominalization by means of a verbal noun, impersonal sentences are examined and typologized.

StructSim: Meta-Structure-Based Similarity Measure in Heterogeneous Information Networks

Similarity measures in heterogeneous information networks (HINs) have become increasingly important in recent years. Most measures in such networks are based on the meta path, a relation sequence connecting object types. However, in real-world scenarios, there exist many complex semantic relationships, which cannot be captured by the meta path. Therefore, a meta structure is proposed, which is a directed acyclic graph of object and relation types. In this paper, we explore the complex semantic meanings in HINs and propose a meta-structure-based similarity measure called StructSim. StructSim models the probability of subgraph expansion with bias from source node to target node. Different from existing methods, StructSim claims that the subgraph expansion is biased, i.e., the probability may be different when expanding from the same node to different nodes with the same type based on the meta structure. Moreover, StructSim defines the expansion bias by considering two types of node information, including out-neighbors of current expanded nodes and in-neighbors of next hop nodes to be expanded. To facilitate the implementation of StructSim, we further designed the node composition operator and expansion probability matrix with bias. Extensive experiments on DBLP and YAGO datasets demonstrate that StructSim is more effective than the state-of-the-art approaches.

Nonfactoid Question Answering as Query-Focused Summarization With Graph-Enhanced Multihop Inference.

Nonfactoid question answering (QA) is one of the most extensive yet challenging applications and research areas in natural language processing (NLP). Existing methods fall short of handling the long-distance and complex semantic relations between the question and the document sentences. In this work, we propose a novel query-focused summarization method, namely a graph-enhanced multihop query-focused summarizer (GMQS), to tackle the nonfactoid QA problem. Specifically, we leverage graph-enhanced reasoning techniques to elaborate the multihop inference process in nonfactoid QA. Three types of graphs with different semantic relations, namely semantic relevance, topical coherence, and coreference linking, are constructed for explicitly capturing the question-document and sentence-sentence interrelationships. Relational graph attention network (RGAT) is then developed to aggregate the multirelational information accordingly. In addition, the proposed method can be adapted to both extractive and abstractive applications as well as be mutually enhanced by joint learning. Experimental results show that the proposed method consistently outperforms both existing extractive and abstractive methods on two nonfactoid QA datasets, WikiHow and PubMedQA, and possesses the capability of performing explainable multihop reasoning.

Graph Learning Augmented Heterogeneous Graph Neural Network for Social Recommendation

Social recommendation based on social network has achieved great success in improving the performance of the recommendation system. Since social network (user-user relations) and user-item interactions are both naturally represented as graph-structured data, Graph Neural Networks (GNNs) have thus been widely applied for social recommendation. Despite the superior performance of existing GNNs-based methods, there are still several severe limitations: (i) Few existing GNNs-based methods have considered a single heterogeneous global graph which takes into account user-user relations, user-item interactions, and item-item similarities simultaneously. That may lead to a lack of complex semantic information and rich topological information when encoding users and items based on GNN. (ii) Furthermore, previous methods tend to overlook the reliability of the original user-user relations which may be noisy and incomplete. (iii) More importantly, the item-item connections established by a few existing methods merely using initial rating attributes or extra attributes (such as category) of items, may be inaccurate or sub-optimal with respect to social recommendation. In order to address these issues, we propose an end-to-end heterogeneous global graph learning framework, namely Graph Learning Augmented Heterogeneous Graph Neural Network (GL-HGNN) for social recommendation. GL-HGNN aims to learn a heterogeneous global graph that makes full use of user-user relations, user-item interactions and item-item similarities in a unified perspective. To this end, we design a Graph Learner (GL) method to learn and optimize user-user and item-item connections separately. Moreover, we employ a Heterogeneous Graph Neural Network (HGNN) to capture the high-order complex semantic relations from our learned heterogeneous global graph. To scale up the computation of graph learning, we further present the Anchor-based Graph Learner (AGL) to reduce computational complexity. Extensive experiments on four real-world datasets demonstrate the effectiveness of our model.

Unified Transformer with Cross-Modal Mixture Experts for Remote-Sensing Visual Question Answering

Remote-sensing visual question answering (RSVQA) aims to provide accurate answers to remote sensing images and their associated questions by leveraging both visual and textual information during the inference process. However, most existing methods ignore the significance of the interaction between visual and language features, which typically adopt simple feature fusion strategies and fail to adequately model cross-modal attention, struggling to capture the complex semantic relationships between questions and images. In this study, we introduce a unified transformer with cross-modal mixture expert (TCMME) model to address the RSVQA problem. Specifically, we utilize the vision transformer (VIT) and BERT to extract visual and language features, respectively. Furthermore, we incorporate cross-modal mixture experts (CMMEs) to facilitate cross-modal representation learning. By leveraging the shared self-attention and cross-modal attention within CMMEs, as well as the modality experts, we effectively capture the intricate interactions between visual and language features and better focus on their complex semantic relationships. Finally, we conduct qualitative and quantitative experiments on two benchmark datasets: RSVQA-LR and RSVQA-HR. The results demonstrate that our proposed method surpasses the current state-of-the-art (SOTA) techniques. Additionally, we perform an extensive analysis to validate the effectiveness of different components in our framework.

Grammatical impairment in schizophrenia: An exploratory study of the pronominal and sentential domains.

Schizophrenia (SZ) is a severe mental disorder associated with a variety of linguistic deficits, and recently it has been suggested that these deficits are caused by an underlying impairment in the ability to build complex syntactic structures and complex semantic relations. Aiming at contributing to determining the specific linguistic profile of SZ, we investigated the usage of pronominal subjects and sentence types in two corpora of oral dream and waking reports produced by speakers with SZ and participants without SZ (NSZ), both native speakers of Brazilian Portuguese. Narratives of 40 adult participants (20 SZ, and 20 NSZ-sample 1), and narratives of 31 teenage participants (11 SZ undergoing first psychotic episode, and 20 NSZ-sample 2) were annotated and statistically analyzed. Overall, narratives of speakers with SZ presented significantly higher rates of matrix sentences, null pronouns-particularly null 3Person referential pronouns-and lower rates of non-anomalous truncated sentences. The high rate of matrix sentences correlated significantly with the total PANSS scores, suggesting an association between the overuse of simple sentences and SZ symptoms in general. In contrast, the high rate of null pronouns correlated significantly with positive PANSS scores, suggesting an association between the overuse of null pronominal forms and the positive symptoms of SZ. Finally, a cross-group analysis between samples 1 and 2 indicated a higher degree of grammatical impairment in speakers with multiple psychotic episodes. Altogether, the results strengthen the notion that deficits at the pronominal and sentential levels constitute a cross-cultural linguistic marker of SZ.

Knowledge and skills extraction from the job requirements texts

The analysis of the requirements for vacancies in the labor market shows that they are multi-level language constructions of several words with complex semantic relationships. The aim of the research is to develop a method for extracting short texts of knowledge and skills from texts of job requirements that have a complex organizational structure. The method consists in supplementing the structure of complex sentences with new relationships by means of a BERT neural network model trained on the texts of online vacancies and moving from a complex text to a set of simple word combinations. The process of additional training (finetunig) of BERT neural network models from the Sberbank AI laboratory on the texts of online vacancies is shown. Two mechanisms for adding new links between requirements words, taking into account knowledge from the subject area, are implemented: linear and through the addition of the parsing tree. In the course of the experiment, a comparative analysis was carried out for several combinations of the listed tools. The combination that showed the best result was 'the BERT + deeppavlov_syntax_parser model + a linear method of adding links'. The applicability of the method was demonstrated on the text corpus of online job requirements. The proposed method has shown higher efficiency than the rule-based approach, which involves the use of formal rules and grammar rules for natural language analysis. Using the method will allow you to quickly identify the key changes in the needs of the labor market at the level of requirements texts of individual knowledge and skills.

Differentiable Meta Multigraph Search with Partial Message Propagation on Heterogeneous Information Networks

Heterogeneous information networks (HINs) are widely employed for describing real-world data with intricate entities and relationships. To automatically utilize their semantic information, graph neural architecture search has recently been developed for various tasks of HINs. Existing works, on the other hand, show weaknesses in instability and inflexibility. To address these issues, we propose a novel method called Partial Message Meta Multigraph search (PMMM) to automatically optimize the neural architecture design on HINs. Specifically, to learn how graph neural networks (GNNs) propagate messages along various types of edges, PMMM adopts an efficient differentiable framework to search for a meaningful meta multigraph, which can capture more flexible and complex semantic relations than a meta graph. The differentiable search typically suffers from performance instability, so we further propose a stable algorithm called partial message search to ensure that the searched meta multigraph consistently surpasses the manually designed meta-structures, i.e., meta-paths. Extensive experiments on six benchmark datasets over two representative tasks, including node classification and recommendation, demonstrate the effectiveness of the proposed method. Our approach outperforms the state-of-the-art heterogeneous GNNs, finds out meaningful meta multigraphs, and is significantly more stable. Our code is available at https://github.com/JHL-HUST/PMMM.

Towards Real-Time Panoptic Narrative Grounding by an End-to-End Grounding Network

Panoptic Narrative Grounding (PNG) is an emerging cross-modal grounding task, which locates the target regions of an image corresponding to the text description. Existing approaches for PNG are mainly based on a two-stage paradigm, which is computationally expensive. In this paper, we propose a one-stage network for real-time PNG, termed End-to-End Panoptic Narrative Grounding network (EPNG), which directly generates masks for referents. Specifically, we propose two innovative designs, i.e., Locality-Perceptive Attention (LPA) and a bidirectional Semantic Alignment Loss (SAL), to properly handle the many-to-many relationship between textual expressions and visual objects. LPA embeds the local spatial priors into attention modeling, i.e., a pixel may belong to multiple masks at different scales, thereby improving segmentation. To help understand the complex semantic relationships, SAL proposes a bidirectional contrastive objective to regularize the semantic consistency inter modalities. Extensive experiments on the PNG benchmark dataset demonstrate the effectiveness and efficiency of our method. Compared to the single-stage baseline, our method achieves a significant improvement of up to 9.4% accuracy. More importantly, our EPNG is 10 times faster than the two-stage model. Meanwhile, the generalization ability of EPNG is also validated by zero-shot experiments on other grounding tasks. The source codes and trained models for all our experiments are publicly available at https://github.com/Mr-Neko/EPNG.git.

Multi-Feature Fusion Method for Chinese Pesticide Named Entity Recognition

Chinese pesticide named-entity recognition (NER) aims to identify named entities related to pesticide properties from unstructured Chinese pesticide information texts. In view of the characteristics of massive, fragmented, professional, and complex semantic relationships of pesticide information data, a deep learning method based on multi-feature fusion was applied to improve the accuracy of pesticide NER. In this study, the pesticide data set is manually annotated by the begin inside outside (BIO) sequence annotation scheme. Bi-directional long short-term memory (BiLSTM) and iterated dilated convolutional neural networks (IDCNN) combined with conditional random field (CRF) form the model BiLSTM-IDCNN-CRF, and it is applied to implement named-entity recognition in Chinese pesticide data sets. IDCNN is introduced to enhance the semantic representation ability and local feature capture ability of the text. BiLSTM network and IDCNN network are combined to obtain the long-distance dependence relationship and context features of different granularity of pesticide data text. Finally, CRF is used to implement the sequence labeling task. According to the experiment results, the accuracy rate, recall rate, and F1 score of the BiLSTM-IDCNN-CRF model in the Chinese pesticide data set were 78.59%, 68.71%, and 73.32%, respectively, which are significantly better than other compared models. Experiments show that the BiLSTM-IDCNN-CRF model can effectively identify and extract entities from Chinese pesticide information text data, which is helpful in constructing the pesticide information knowledge graph and intelligent question-answering.

Document-level relation extraction with two-stage dynamic graph attention networks

Document-level Relation Extraction (RE) aims to infer complex semantic relations between entities in a document. Previous approaches leverage a multi-classification model to predict relation types between each entity pair. However, in contrast to sentence-level RE, document-level RE contains various entities expressed by mentions appearing across multiple sentences in a document. Therefore, the amount of negative instances (‘no relationship’) significantly outnumbers that of other positive instances in document-level RE. In addition, most existing methods construct static graphs with heuristic rules to capture the interactions among entities. However, these heuristic rules ignore the specificities of the documents. In this study, we propose a novel two-stage framework to extract document-level relations based on dynamic graph attention networks, namely TDGAT. In the first stage, we capture the relational links of the entity pairs using a binary classification model. In the second stage, we extract fine-grained relations among entities, including the type of ‘NA (no relationship)’. To reduce error propagation, we regard the entity pair links predicted in the first stage as the prior information and leverage them to reconstruct the document-level graphs of the second stage. In this manner, we can provide extra head and tail entity connection information for predicting relations in a document. Furthermore, we propose a dynamic graph strategy to explore the multi-hop interactions between related information. The experimental results show that our framework outperforms most existing models on the public document-level dataset DocRED. The extensive analysis demonstrates the effectiveness of our TDGAT in extracting inter-sentence relations.

A study on pharmaceutical text relationship extraction based on heterogeneous graph neural networks.

Effective information extraction of pharmaceutical texts is of great significance for clinical research. The ancient Chinese medicine text has streamlined sentences and complex semantic relationships, and the textual relationships may exist between heterogeneous entities. The current mainstream relationship extraction model does not take into account the associations between entities and relationships when extracting, resulting in insufficient semantic information to form an effective structured representation. In this paper, we propose a heterogeneous graph neural network relationship extraction model adapted to traditional Chinese medicine (TCM) text. First, the given sentence and predefined relationships are embedded by bidirectional encoder representation from transformers (BERT fine-tuned) word embedding as model input. Second, a heterogeneous graph network is constructed to associate words, phrases, and relationship nodes to obtain the hidden layer representation. Then, in the decoding stage, two-stage subject-object entity identification method is adopted, and the identifier adopts a binary classifier to locate the start and end positions of the TCM entities, identifying all the subject-object entities in the sentence, and finally forming the TCM entity relationship group. Through the experiments on the TCM relationship extraction dataset, the results show that the precision value of the heterogeneous graph neural network embedded with BERT is 86.99% and the F1 value reaches 87.40%, which is improved by 8.83% and 10.21% compared with the relationship extraction models CNN, Bert-CNN, and Graph LSTM.

Knowledge graph-enabled adaptive work packaging approach in modular construction

Adaptive work packaging is paramount in helping reduce dynamic gaps between design and manufacturing in modular construction (MC), particularly in mass customization. However, current work packaging methods fail to automatically extract complex semantic relations among work package elements (e.g., products, tasks, and their dependencies) and dynamically reason the implicit semantic knowledge (e.g., the different granularity of semantics) as the project progresses. To address these issues, this study proposes a knowledge graph-enabled adaptive work packaging (K-GAWP) approach to dynamically form semantic-enriched work packages with different granularities. Thus far, this study first models the data of tasks, products, and their spatial relationships for MC production as graphs. Second, a novel multi-granularity knowledge reasoning method (product2task) is developed to map products to tasks in an adaptive manner. Third, a dedicated hierarchical clustering method (task2package) involving multiple features from the dependency structure matrix is proposed for work-package generation (i.e., task knowledge fusion). Finally, the K-GAWP’s performance is evaluated through controlled experiments in a real MC project. The results indicate that the K-GAWP approach performs work packaging in an adaptive, accurate, and efficient manner, thereby improving the distributed planning and control of MC projects.

Ontology-Based Relational Product Recommendation System.

As online shopping has expanded, product recommendations on e-commerce websites have gained significance. Systems for recommending products use information about site navigation and user leave-over to suggest more products. Customers who use a product recommendation system choose better and find items more quickly. On e-commerce websites, collaborative and content-based filtering is used in product suggestion algorithms. Collaborative filtering is driven by user preference similarity and content-based filtering. While content-based filtering groups are related to products, collaboration groups are like-minded individuals. In collaborative filtering, users with similar user profiles are used during the proposal phase; in content-based filtering, users with similar product profiles are found and recommended. These techniques cannot deliver complex commodities and have slow start-up times and small element sets. Users can push the same product if they only like certain things, but they cannot recommend a new product or user who just joined the system because they are not a group member. These approaches cannot capture complex semantic relationships, making them inadequate for recommending complex products. Recent research has focused on incorporating the domain ontology into the proposition process to create a more precise and helpful suggestion. The relational qualities of the product are not covered in this study, only its category and features are. Actually, the ontology of the proposed product should be included in the suggestion system. Relational data is integrated into the recommendation engine in this study using domain ontologies. This was done to research books that people had recommended. Relational data from an online bookseller was used to test the proposed infrastructure.

Rethinking the Two-Stage Framework for Grounded Situation Recognition

Grounded Situation Recognition (GSR), i.e., recognizing the salient activity (or verb) category in an image (e.g.,buying) and detecting all corresponding semantic roles (e.g.,agent and goods), is an essential step towards “human-like” event understanding. Since each verb is associated with a specific set of semantic roles, all existing GSR methods resort to a two-stage framework: predicting the verb in the first stage and detecting the semantic roles in the second stage. However, there are obvious drawbacks in both stages: 1) The widely-used cross-entropy (XE) loss for object recognition is insufficient in verb classification due to the large intra-class variation and high inter-class similarity among daily activities. 2) All semantic roles are detected in an autoregressive manner, which fails to model the complex semantic relations between different roles. To this end, we propose a novel SituFormerfor GSR which consists of a Coarse-to-Fine Verb Model (CFVM) and a Transformer-based Noun Model (TNM). CFVM is a two-step verb prediction model: a coarse-grained model trained with XE loss first proposes a set of verb candidates, and then a fine-grained model trained with triplet loss re-ranks these candidates with enhanced verb features (not only separable but also discriminative). TNM is a transformer-based semantic role detection model, which detects all roles parallelly. Owing to the global relation modeling ability and flexibility of the transformer decoder, TNM can fully explore the statistical dependency of the roles. Extensive validations on the challenging SWiG benchmark show that SituFormer achieves a new state-of-the-art performance with significant gains under various metrics. Code is available at https://github.com/kellyiss/SituFormer.

Dual-Channel and Hierarchical Graph Convolutional Networks for document-level relation extraction

Document-level relation extraction aims to infer complex semantic relations among entities in an entire document. Compared with the sentence-level relation extraction, document-level relational facts are expressed by multiple mentions across the sentences in a long-distance, requiring excellent reasoning. In this paper, we propose Dual-Channel and Hierarchical Graph Convolutional Networks (DHGCN), which constructs three graphs in token-level, mention-level, and entity-level to model complex interactions among different semantic representations across the document. Based on the multi-level graphs, we apply the Graph Convolutional Network (GCN) for each level to aggregate the relevant information scattered throughout the document for better inferring the implicit relations. Moreover, we propose a dual-channel encoder to capture structural and contextual information simultaneously, which also supplies the contextual representation for the higher layer to avoid losing low-dimension information. Our DHGCN yields significant improvements over the state-of-the-art methods by 2.75, 5.5, and 3.5 F1 on DocRED, CDR, and GDA, respectively, which are popular document-level relation extraction datasets. Furthermore, to demonstrate the effectiveness of our method, we evaluate DHGCN on a fine gained clinical document-level dataset Symptom-Acupoint Relation (SAR) proposed by ourselves and available at https://github.com/QiSun123/SAR. The experimental results illustrate that DHGCN is able to infer more valuable relations among entities in the document.

The conceptual semantics of “money” and “money verbs”

The central purpose of this study is to apply the NSM (Natural Semantic Metalanguage) method of semantic-conceptual analysis to the word money and to related economic transaction verbs, such as buy, sell and pay, as used in everyday English. It proposes semantic explications for these words on the basis of conceptual analysis and a range of linguistic evidence and taking account of lexical polysemy. Even in its basic meaning (in a sentence like there was some money on the table), money-1 is shown to be surprisingly complex, comprising about 35 lines of semantic text and drawing on a number of semantic molecules (such as country, number, and hands), as well as a rich assortment of semantic primes. This money-1 meaning turns out to be a crucial semantic molecule in the composition of the verbs buy, sell, pay, and (it) costs. Each of these is treated in some detail, thereby bringing to light the complex semantic relationships between them and clarifying how this bears on their grammatical properties, such as argument structure. The concluding section considers how NSM semantic-conceptual analysis can help illuminate everyday economic thinking and also how it connects with Humanonics, an interdisciplinary project which aims to re-humanise economics.

Solving Sensor Ontology Metamatching Problem with Compact Flower Pollination Algorithm

To implement co-operation among applications on the Internet of Things (IoT), we need to describe the meaning of diverse sensor data with the sensor ontology. However, there exists a heterogeneity issue among different sensor ontologies, which hampers their communications. Sensor ontology matching is a feasible solution to this problem, which is able to map the identical ontology entity pairs. This work investigates the sensor ontology meta-matching problem, which indirectly optimizes the sensor ontology alignment’s quality by tuning the weights to aggregate different ontology matchers. Due to the largescale entity and their complex semantic relationships, swarm intelligence (SI) based techniques are emerging as a popular approach to optimize the sensor ontology alignment. Inspired by the success of the flower pollination algorithm (FPA) in the IoT domain, this work further proposes a compact FPA (CFPA), which introduces the compact encoding mechanism to improve the algorithm’s efficiency, and on this basis, the compact exploration and exploitation operators are proposed, and an adaptive switching probability is presented to trade-off these two searching strategies. The experiment uses the ontology alignment evaluation initiative (OAEI)’s benchmark and the real sensor ontologies to test CFPA’s performance. The statistical comparisons show that CFPA significantly outperforms other state-of-the-art sensor ontology matching techniques.

Fact-Driven Abstractive Summarization by Utilizing Multi-Granular Multi-Relational Knowledge

Abstractive summarization generates a concise summary to capture the key ideas of the source text. This task underpins important applications like information retrieval, document comprehension, and event tracking. While much progress has been achieved, state-of-the-art summarization approaches often fail to generate high-quality summaries to reproduce factual details accurately. One of the key limitations of existing solutions is that they are primarily concerned about extracting facts from the source text but overlook other crucial factual information, such as the related time, locations, reasons, consequences, purposes, participants and involved parties. Furthermore, the current summarization frameworks are inadequate in modeling the complex semantic relations among facts and the corresponding factual information, leaving much room for improvement. This paper presents <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">FFSum</small> , a novel summarization framework for exploiting multi-grained factual information to improve text summarization. To this end, <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">FFSum</small> constructs an individual fine-grained factual graph with multiple relations among facts and the corresponding factual information. It employs a fact-driven graph attention network to integrate multi-granular factual representations at the encoding stage. It then uses a hybrid pointer network to retrieve factual pieces from the graph for the summary generation. We evaluate the <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">FFSum</small> by applying it to two real-world datasets. Experimental results show that the <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">FFSum</small> consistently outperforms a state-of-the-art approach across evaluation datasets.