Indexing device and method based on time sequence core subgraph query

By constructing directed acyclic graphs and super-optimized indexes, the dominant properties of α and β values ​​are used to solve the high cost and complexity problems of timing (α,β)-core subgraph query, and the efficiency of dynamic time graphs is achieved Query and effectiveness improvement.

CN118939846APending Publication Date: 2024-11-12ZHEJIANG CHUANGLIN TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410986040.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The existing timing (α,β)-core subgraph query methods have high cost, spatial complexity and lack of effectiveness, especially when dealing with dynamic time graphs, static solutions cannot adapt to time window changes, online solutions The stripping process is expensive and the timestamp information cannot be recognized.

Method used

By constructing a directed acyclic graph based on a time series binary graph, using the dominant attributes of α and β values, a super-optimized index is built, spatial complexity is optimized, and query performance is improved. Obsolete information is removed through maintenance strategies and included in the newly arrived time edge , suitable for dynamic time graphs.

Benefits of technology

The spatial complexity and query performance are significantly optimized, the index effectiveness is improved, and it is suitable for processing real-world dynamic time graphs, and efficient query of core subgraphs is realized.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118939846A_ABST
    Figure CN118939846A_ABST
Patent Text Reader

Abstract

The embodiment of the invention discloses an indexing device and method based on time sequence core subgraph query, and the device comprises a construction module which is used for constructing a directed acyclic graph containing a core pair (alpha, beta) based on an obtained time sequence bipartite graph; the directed acyclic graph has a qualified time window; constructing a vertex-based index based on the time window; the optimization module is used for constructing a super optimization index according to the index of the vertex; updating the super optimization index based on a preset maintenance strategy; the processing module is used for querying the structure of the core sub-graph according to the updated super optimization index so as to obtain a corresponding core sub-graph result; the method has the advantages that the super optimization index is constructed, the space complexity is remarkably optimized, the query performance is improved, and the effectiveness is improved by removing outdated information and incorporating the outdated information into a new arrival time edge.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of information technology technology, and in particular to an indexing device and method based on time-series core subgraph query. Background Art

[0002] Time-series graph query: A technique for querying time-series graph data that allows users to query graph data based on timestamps. This technique is very useful when processing graph data with timestamps, such as dynamic relationships in social networks, attack patterns in network security, etc. Time-series graph query can help users discover patterns and trends in time-series data, allowing for deeper analysis and decision-making.

[0003] The (α,β)-core subgraph is a dense subgraph model in a bipartite graph, where α represents the minimum degree requirement of the upper nodes in the bipartite graph, and β represents the minimum degree requirement of the lower nodes in the bipartite graph.

[0004] At present, there are many works exploring various types of time core class subgraph query research. There are also many queries based on time windows, including historical k-core, and many works focusing on the research of core class structure in persistent communities in time graphs, online method solutions and static solutions; however, the above solutions have the following defects in solving the problem of temporal (α, β)-core subgraph query:

[0005] 1. The online solution requires a peeling process for each temporal (α, β)-core query; for the solution to the persistent (α, β)-core community search problem (M.Li, Z.Xie, and L.Ding. Persistent community search over temporal bipartite graphs. In ADMA, pages 324-339. Springer, 2023), it uses (α, β)-cores to find persistent communities for a given query vertex, where the community persists the (α, β)-core structure for a given time interval; specifically, this method first performs a graph reduction based on the given query vertex, which makes it unsuitable for the temporal (α, β)-core query problem. Although a simple online method can be derived from this method, the peeling process for each query will incur a high cost, which is unacceptable in practical applications;

[0006] 2. Existing static solutions are not feasible for solving this timing problem due to high space and construction costs. Considering the existing static (α, β)-core query solution Bicore-Index (B.Liu, L.Yuan, X.Lin, L.Qin, W.Zhang, and J.Zhou. Efficient (α, β)-core computation: An index-based approach. WWW, pages 1130-1141, 2019), Bicore-Index cannot recognize timestamp information; although the time version of Bicore-Index can be derived by building an index for each snapshot in any possible time window; however, the construction of this method will incur time complexity and space complexity;

[0007] 3. Lack of effectiveness: Since the (α, β)-core has two related parameters that need to be considered simultaneously, where α and β represent the threshold requirements of the two-layer vertices; therefore, a two-dimensional method rather than one-dimensional monotonicity is needed to describe the relationship between the (α, β)-core, which excludes the feasibility of the historical k-core solution; in addition, as the time window changes, old edges are constantly excluded and new edges are constantly considered; this causes the (α, β)-core to constantly change, making the simple inclusion rule applicable in static scenarios no longer applicable to the dynamic properties of real-world time graphs. Summary of the invention

[0008] In response to the technical defects mentioned in the background technology, the purpose of an embodiment of the present invention is to provide an indexing device and method based on time-series core subgraph query, which constructs a super-optimized index by utilizing the dominant attribute between α and β values, significantly optimizes the space complexity, improves the query performance, and improves the effectiveness by removing outdated information and incorporating newly arrived time edges.

[0009] To achieve the above objectives, in a first aspect, an embodiment of the present invention provides an indexing device based on time-series core subgraph query, the device comprising:

[0010] Building blocks for:

[0011] Based on the obtained time series bipartite graph, a directed acyclic graph including core pairs (α, β) is constructed to describe the inclusion relationship between all (α, β)-cores; wherein the directed acyclic graph has a qualified time window;

[0012] Build vertex-based indexes based on time windows;

[0013] Optimization modules for:

[0014] Constructing a super optimized index according to the index of the vertex;

[0015] Updating the super optimization index based on a preset maintenance strategy;

[0016] The processing module is used to query the structure of the core subgraph according to the updated super-optimized index to obtain the corresponding core subgraph result.

[0017] As a specific implementation of the present application, the construction of the vertex-based index specifically includes:

[0018] Calculate all core subgraphs of the inputted temporal bipartite graph in a preset first range of time;

[0019] Initialize all auxiliary structures;

[0020] Determine whether time t is within the set first time interval; if so, delete all edges at time t; and maintain the core value of the edge under the preset first time window;

[0021] If not, the start time is maintained in the preset second time window, and a vertex-based index is constructed for output.

[0022] As a specific implementation of the present application, the super optimized index specifically includes:

[0023] The dominant core values ​​corresponding to all qualified time windows of all nodes are calculated and linked to output the super optimization index.

[0024] As a specific implementation of the present application, the updating of the super optimization index based on a preset maintenance strategy specifically includes:

[0025] Obtaining the stale timestamp of the existing super-optimized index and the new time window;

[0026] Delete the parts related to the old time window in all sub-indexes; and keep the parts related to the time interval between the old timestamp and the new time window;

[0027] Update the impact of the edges in the new time window on the original core value;

[0028] Updates to qualifying time windows and associated ancillary structures;

[0029] Outputs the updated super-optimized index.

[0030] As a specific implementation of the present application, the vertex-based index is composed of sub-indexes of all vertices, wherein the sub-indexes are all qualified time windows sorted for each core pair and follow a DAG-like hierarchy.

[0031] In a second aspect, an embodiment of the present invention further provides an indexing method based on time-series core subgraph query, the method comprising the following steps:

[0032] Based on the obtained time series bipartite graph, a directed acyclic graph including core pairs (α, β) is constructed to describe the inclusion relationship between all (α, β)-cores; wherein the directed acyclic graph has a qualified time window;

[0033] Build vertex-based indexes based on time windows;

[0034] Constructing a super optimized index according to the index of the vertex;

[0035] Updating the super optimization index based on a preset maintenance strategy;

[0036] The structure of the core subgraph is queried according to the updated super-optimized index to obtain the corresponding core subgraph result.

[0037] The technical solution provided by the embodiment of the present invention proposes a novel hierarchical structure similar to a directed acyclic graph (DAG) with a qualified time window to describe the time inclusion property of the (α, β)-core; utilizes the dominant attribute between the α and β values ​​to construct a super-optimized index that significantly optimizes the space complexity while ensuring efficient query performance; and based on a maintenance strategy, by removing outdated information and incorporating newly arrived time edges, the effectiveness is improved and the dynamic properties suitable for processing real-world time graphs are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the specific implementation of the present invention or the technical solution in the prior art, the drawings required for use in the specific implementation or the description of the prior art are briefly introduced below.

[0039] Figure 1 It is a structural diagram of an indexing device based on time-series core subgraph query provided by an embodiment of the present invention;

[0040] Figure 2 It is a hierarchical structure diagram of a directed acyclic graph provided by an embodiment of the present invention;

[0041] Figure 3 This is a vertex-based index construction process provided by an embodiment of the present invention;

[0042] Figure 4 This is a processing flow of a super-optimized index provided by an embodiment of the present invention;

[0043] Figure 5 This is a processing flow for updating a super optimized index provided by an embodiment of the present invention;

[0044] Figure 6 It is a flow chart of an indexing method based on time-series core subgraph query provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0047] Please refer to Figures 1 to 5 , an indexing device based on time-series core subgraph query provided by an embodiment of the present invention comprises:

[0048] Building blocks for:

[0049] Based on the obtained time series bipartite graph, a directed acyclic graph including core pairs (α, β) is constructed to describe the inclusion relationship between all (α, β)-cores; wherein the directed acyclic graph has a qualified time window;

[0050] Build vertex-based indexes based on time windows;

[0051] Optimization modules for:

[0052] Constructing a super optimized index according to the index of the vertex;

[0053] Updating the super optimization index based on a preset maintenance strategy;

[0054] The processing module is used to query the structure of the core subgraph according to the updated super-optimized index to obtain the corresponding core subgraph result.

[0055] It should be noted that, in this embodiment, (α, β)-core and (α, β)-core both refer to dense subgraphs, and the minimum degree values ​​of the nodes in the two-layer graphs are α and β respectively; the core pair (α, β) refers to a pair of values ​​(α, β).

[0056] In this embodiment, refer to Figure 2, a directed acyclic graph (DAG)-like hierarchical structure is constructed to describe the inclusion relationship between all possible (α, β)-cores, where αm represents the maximum α value, βm represents the maximum β value, and δ represents the maximum value of the (α, β)-core existence such that α=β=δ.

[0057] To obtain (α, β)-coreness under different time windows, a more efficient approach is to record the time of each vertex change compared to the simple index solution of computing all possible windows. From this perspective, the concept of shortest (α, β)-coreness time is predefined: given a temporal bipartite graph G, two positive integers α and β, a start time ts and a vertex u∈V(G), the shortest (α, β)-coreness time of u under ts and (α, β) is the earliest time te such that u is in S<te,te> In the (α, β)-core inside, it is denoted as ST(u,(α,β),ts).

[0058] The concept of qualified time window is proposed: given a temporal bipartite graph G, two positive integers α and β, a start time ts, a vertex u∈V(G) and its shortest (α, β)-core time ST(u,(α,β),ts), a qualified time window [t's,t'e] is a qualified time window if and only if the following two conditions are met:

[0059] 1) t'e = ST(u, (α, β), ts), that is, the end time of the time window is equal to the shortest (α, β)-core time of vertex u;

[0060] 2) t's is the minimum time such that ST(u,(α,β),t's)=t'e, that is, the start time of the time window is the minimum time such that the shortest (α,β)-core time of u is equal to the end time of the window t'e.

[0061] Then QTW(u,(α,β)) represents all qualified windows of u for the (α,β)-core.

[0062] By combining a DAG-like hierarchy and qualified time windows, we can build an index for each vertex to answer temporal (α, β)-core queries.

[0063] The vertex-based index consists of sub-indexes of all vertices, where the sub-indexes are all eligible time windows sorted for each core pair and follow a DAG-like hierarchy.

[0064] Further, refer to Figure 3 , the construction of the vertex-based index specifically includes:

[0065] All core subgraphs are calculated for the inputted temporal bipartite graph in a preset first range of time; wherein the first range of time is Figure 3 <1, tmax> time snapshot in;

[0066] Initialize all auxiliary structures;

[0067] Determine whether time t is within the set first time interval; if so, delete all edges of time t; and maintain the core value of the edge under the preset first time window; where the first time interval is Figure 3 [tmax,2] in the first time window is <1, t-1>;

[0068] If not, the maintenance start time is in the preset second time window, and a vertex-based index is constructed for output; wherein the second time window is Figure 3 〈2, tmax> in.

[0069] tmax refers to the latest timestamp that can be obtained in the current timing diagram; boundary 1 in <1, tmas> refers to the initial normalized timestamp; boundary 2 in <2, tmas> refers to a continuous, non-missing time window;

[0070] The auxiliary structures in this article refer to all the data structures required to execute the algorithm, which are used to store and calculate intermediate information;

[0071] The core value is a core pair (α, β), which refers to a pair of values ​​(α, β).

[0072] In this embodiment, given a correctly constructed vertex-based index (denoted by IV) and a time (α, β)-core query, the query processing of IV starts by traversing all vertices in G. For each sub-index IV(u), each core pair is set as a query entry; if there is no entry (α, β) for this query, then it is an invalid query and the correct result should be an empty set; after the target core pair entry matches the query (α, β), the process of checking the qualified time window is implemented by binary search.

[0073] Reference Figure 4 , the super optimized index specifically includes:

[0074] The dominant core values ​​corresponding to all qualified time windows of all nodes are calculated and linked to output the super optimization index.

[0075] Specifically, the dominant core values ​​corresponding to all qualified time windows of all nodes are calculated and linked, as follows:

[0076] Based on the given sorted core pairs, we need to perform two rounds of filtering on the two layers of vertices in the bipartite graph respectively; we filter out the vertices that are not the dominant core pairs along the α direction, and we perform similar filtering on the lower layer vertices; after two rounds of filtering, we obtain the dominant core value of each vertex in a specific qualified time window (that is, the core pair mentioned above).

[0077] For the linking rule: Consider two core pairs (α, β) and (α', β'). If the Manhattan distance between them is exactly 1, they are said to be continuous. In real-world bipartite graphs, core pairs may not be continuous in the DAG, so we need to deal with this discontinuity, which is necessary to ensure correct linking. In general, for a dominant core value (α, β), the linking rule with another dominant core value (α', β') is as follows:

[0078] Rule 1. Direct link: If (α, β) has the same value as (α', β') on α (or β), that is, α=α' (or β=β'), then when the difference in the other dimension t=|β'-β| (or t=|α'-α|) is minimized, (α, β) and (α', β') can be directly linked.

[0079] Rule 2. Skip Linking: If (α,β) does not share the same value with (α',β') on both α and β, then they can be skip linked only if the Manhattan distance between (α,β) and (α',β') is the smallest among all possible (α',β').

[0080] Reference Figure 5 , the super optimization index is updated based on a preset maintenance strategy, specifically including:

[0081] Obtaining the stale timestamp of the existing super-optimized index and the new time window;

[0082] Delete the parts related to the old time window in all sub-indexes; and keep the parts related to the time interval between the old timestamp and the new time window;

[0083] Update the impact of the edges in the new time window on the original core value;

[0084] Updates to qualifying time windows and associated ancillary structures;

[0085] Outputs the updated super-optimized index.

[0086] Reference Figure 5 , in this embodiment, the stale timestamp is represented by tsta;

[0087] The new time window is<tmax+1,tmax′> Indicates that the stale time window is <1, tsta>;

[0088] The time interval between the old timestamp and the new time window is<tsta+1,tamx> .

[0089] It should be noted that tamx′ has the same meaning as tmax′ in the figure, both referring to the latest timestamp after the update time;

[0090] During maintenance, delete the references related to the obsolete time window in all sub-indexes. The edge under the time window composed of the right boundary tsta of the obsolete information time and the left boundary 1 of the oldest time, the part they affect in the index structure needs to be deleted;

[0091] The relevant part of the time interval between the old timestamp and the new time window is referred to as the edge under the time window consisting of the left boundary tsta+1 to the new right boundary tmax′.

[0092] For example, it is necessary to discard outdated edges within <1, tsta and merge new edges within tmax+1, t′max, where t′max represents the new latest timestamp, tsta represents the right boundary of the outdated time window that needs to be discarded, and tmax+1 means the left boundary of the time window that needs to be newly supplemented.

[0093] For super-optimized indexes, removing obsolete information is straightforward and can be done by removing its eligible time windows ts∈<1, tsta>. However, handling new edges inserted within tmax+1, t′max remains a challenge.

[0094] The strategy adopted by this application is: first discard the outdated part of the existing index (i.e., the part related to the outdated edges within the <1, tsta> time window), and then it retains all snapshots S<tsta+1,tmax> The (α, β)-core of the<tsta+1,tmax> And insert edge E(S<tsta+1,t′max> ) Get snapshot S<tsta+1,t′max> Note that the qualified time window and the dominant core center are updated only when the new edge affects the qualified time window, and in the worst case, the subgraph visited in the process is usually much smaller than the entire time graph.

[0095] From the above description, it can be seen that the technical solution provided by the embodiment of the present invention proposes a novel hierarchical structure similar to a directed acyclic graph (DAG) with a qualified time window to describe the time inclusion property of the (α, β)-core; utilizes the dominant attribute between the α and β values ​​to construct a super-optimized index that significantly optimizes the space complexity while ensuring efficient query performance; and based on the maintenance strategy, by removing outdated information and incorporating newly arrived time edges, the effectiveness is improved and the dynamic properties suitable for processing real-world time graphs are achieved.

[0096] like Figure 6 As shown, an embodiment of the present invention further provides an indexing method based on time series core subgraph query, the method comprising the following steps:

[0097] Based on the obtained time series bipartite graph, a directed acyclic graph including core pairs (α, β) is constructed to describe the inclusion relationship between all (α, β)-cores; wherein the directed acyclic graph has a qualified time window;

[0098] Build vertex-based indexes based on time windows;

[0099] A super optimized index is constructed according to the index of the vertex; that is, Figure 6 In, optimize the existing index to a super-optimized index;

[0100] The super-optimization index is updated based on a preset maintenance strategy; that is, Figure 6 In , a super optimized index is maintained based on old and updated information;

[0101] The structure of the core subgraph is queried according to the updated super-optimized index to obtain the corresponding core subgraph result; that is, Figure 6 In , the structure of the core subgraph is queried according to the index, and the core subgraph result is output.

[0102] It should be noted that for a more specific description of the workflow of the method embodiment, please refer to the aforementioned device embodiment part, which will not be repeated here.

[0103] Furthermore, the construction of the vertex-based index specifically includes:

[0104] Calculate all core subgraphs of the inputted temporal bipartite graph in a preset first range of time;

[0105] Initialize all auxiliary structures;

[0106] Determine whether time t is within the set first time interval; if so, delete all edges at time t; and maintain the core value of the edge under the preset first time window;

[0107] If not, the start time is maintained in the preset second time window, and a vertex-based index is constructed for output.

[0108] The super optimized index specifically includes:

[0109] The dominant core values ​​corresponding to all qualified time windows of all nodes are calculated and linked to output the super optimization index.

[0110] Furthermore, the updating of the super optimization index based on a preset maintenance strategy specifically includes:

[0111] Obtaining the stale timestamp of the existing super-optimized index and the new time window;

[0112] Delete the parts related to the old time window in all sub-indexes; and keep the parts related to the time interval between the old timestamp and the new time window;

[0113] Update the impact of the edges in the new time window on the original core value;

[0114] Updates to qualifying time windows and associated ancillary structures;

[0115] Outputs the updated super-optimized index.

[0116] The entire program studies the<ts,te> A new problem of querying within (α, β)-cores is proposed to help find time-sensitive results in real-world applications. A novel directed acyclic graph (DAG)-like hierarchy with qualified time windows is proposed. This hierarchy describes the temporal inclusion property of (α, β)-cores, and the related vertex-based index is effective for solving this problem. By exploiting the dominant property between α and β values, a super-optimized index is constructed, which significantly optimizes the space complexity while ensuring efficient query performance. In order to handle the dynamic properties of real-world temporal graphs, an efficient maintenance algorithm is proposed to improve the effectiveness by removing outdated information and incorporating newly arrived temporal edges.

[0117] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. An indexing device based on time-series core subgraph query, It is characterized in that The device comprises: Building blocks for: Based on the obtained time series bipartite graph, a directed acyclic graph including core pairs (α, β) is constructed to describe the inclusion relationship between all (α, β)-cores; wherein the directed acyclic graph has a qualified time window; Build vertex-based indexes based on time windows; Optimization modules for: Constructing a super optimized index according to the index of the vertex; Updating the super optimization index based on a preset maintenance strategy; The processing module is used to query the structure of the core subgraph according to the updated super-optimized index to obtain the corresponding core subgraph result.

2. An indexing device based on time-series core subgraph query as claimed in claim 1, It is characterized in that The construction of the vertex-based index specifically includes: Calculate all core subgraphs of the inputted temporal bipartite graph in a preset first range of time; Initialize all auxiliary structures; Determine whether time t is within the set first time interval; if so, delete all edges at time t; and maintain the core value of the edge under the preset first time window; If not, the start time is maintained in the preset second time window, and a vertex-based index is constructed for output.

3. An indexing device based on time-series core subgraph query as claimed in claim 1, It is characterized in that The super optimized index specifically includes: The dominant core values ​​corresponding to all qualified time windows of all nodes are calculated and linked to output the super optimization index.

4. An indexing device based on time-series core subgraph query as claimed in claim 1, It is characterized in that The updating of the super optimization index based on the preset maintenance strategy specifically includes: Obtaining the stale timestamp of the existing super-optimized index and the new time window; Delete the parts related to the old time window in all sub-indexes; and keep the parts related to the time interval between the old timestamp and the new time window; Update the impact of the edges in the new time window on the original core value; Updates to qualifying time windows and associated ancillary structures; Outputs the updated super-optimized index.

5. An indexing device based on time-series core subgraph query as claimed in claim 2, It is characterized in that The vertex-based index consists of sub-indexes of all vertices, where the sub-indexes are all eligible time windows sorted for each core pair and follow a DAG-like hierarchy.

6. An indexing method based on time-series core subgraph query, It is characterized in that The method comprises the following steps: Based on the obtained time series bipartite graph, a directed acyclic graph including core pairs (α, β) is constructed to describe the inclusion relationship between all (α, β)-cores; wherein the directed acyclic graph has a qualified time window; Build vertex-based indexes based on time windows; Constructing a super optimized index according to the index of the vertex; Updating the super optimization index based on a preset maintenance strategy; The structure of the core subgraph is queried according to the updated super-optimized index to obtain the corresponding core subgraph result.

7. An indexing method based on time-series core subgraph query as claimed in claim 6, It is characterized in that The construction of the vertex-based index specifically includes: Calculate all core subgraphs of the inputted temporal bipartite graph in a preset first range of time; Initialize all auxiliary structures; Determine whether time t is within the set first time interval; if so, delete all edges at time t; and maintain the core value of the edge under the preset first time window; If not, the start time is maintained in the preset second time window, and a vertex-based index is constructed for output.

8. An indexing method based on time-series core subgraph query as claimed in claim 6, It is characterized in that The super optimized index specifically includes: The dominant core values ​​corresponding to all qualified time windows of all nodes are calculated and linked to output the super optimization index.

9. The indexing method based on time-series core subgraph query according to claim 7, It is characterized in that The updating of the super optimization index based on the preset maintenance strategy specifically includes: Obtaining the stale timestamp of the existing super-optimized index and the new time window; Delete the parts related to the old time window in all sub-indexes; and keep the parts related to the time interval between the old timestamp and the new time window; Update the impact of the edges in the new time window on the original core value; Updates to qualifying time windows and associated ancillary structures; Outputs the updated super-optimized index.