CBPO

Monthly Archives: August 2020

Entity Seeking Queries and Semantic Dependency Trees

August 31, 2020 No Comments

Entity Seeking Queries and Semantic Dependency Trees

Queries for some searches may be entity seeking queries.

Someone may ask, “What is the hotel that looks like a sail.” That query may look for an entity such as the building, the Burj Al Arab Jumeirah.

That entity may be identified by looking at Semantic Dependency Trees, which answer a question in the query (example below)

Other queries are not entity seeking queries and do not look for answers about specific entities. An example is “What is the weather today?” An answer to a query like that could be, “The weather will be between 60-70 degrees Fahrenheit, and sunny today.”

Actions May Accompany Queries that Seek Entities

Google was granted a patent on answering entity seeking queries.

The process in the patent may perform particular actions for queries seeking one or more entities.

Those actions may include:

  • Identifying one or more entities a query is seeking
  • Determining whether the query seeks one specific entity or more than one

For example, the patent may decide a query “What is the hotel that looks like a sail” is looking for a single entity that is a hotel.

In another example, a query “What restaurants nearby serve omelets” looks for many entities that are restaurants.

Alternatively, the system may perform may find a most relevant entity or entities, and present what is identified to a searcher if it is sufficiently relevant to the query. For example, it may identify that the Burj Al Arab Jumeirah is a hotel and is relevant to the term “looks like a sail,” and, in response, audibly output synthesized speech of “Burj Al Arab Jumeirah.”

Additional Dialog about a Query to Concatenate an Entity Seeking Query

Yet another addition or alternative action may include initiating a dialog with the user for more details about the entities that are sought.

For example, the system may determine that a query is seeking a restaurant and there may be two entities that are restaurants are very relevant to the terms in the query and, in response, ask the searcher “Can you give me more details” and concatenate additional input from the user to the original query and re-execute the concatenated query.

Identifying SubQueries of Entity Seeking Queries

Another additional or alternative action may include identifying subqueries of a query which are entity-seeking, and using the above actions to answer the subquery, and then replacing the subqueries by their answers in the original query to obtain a partially resolved query which can be executed.

For example, the system may receive a query of “Call the hotel that looks like a sail,” determine that “the hotel that looks like a sail” is a subquery that seeks an entity, determine an answer to the subquery is “Burj Al Arab Jumeirah,” in response replace “the hotel that looks like a sail” in the query with “The Burj Al Arab Jumeirah” to obtain a partially resolved query of “Call the Burj Al Arab Jumeirah,” and then executes the partially resolved query.

Looking at Previous Queries

Another additional or alternative action may include identifying that a user is seeking entities and adapting how the system resolve queries accordingly.

For example, the system may determine that sixty percent of the previous five queries that a user searched for in the past two minutes sought entities and, in response, determine that a next query that a user provides is more likely an entity seeking query, and process the query accordingly.

An Advantage From Following this Process

An advantage may be more quickly resolving queries in a manner that satisfies a searcher.

For example, the system may be able to immediately provide an actual answer of “The Burj Al Arab Jumeirah” for the query “What hotel looks like a sail” where another system may instead provide a response of “no results found” or provide a response that is a search result listing for the query.

Entity Seeking Queries and Semantic Dependency Trees

Entity Seeking Queries
Another advantage may be that the process may be able to more efficiently identify an entity sought by a query. For example, it may determine an entity seeking query is looking for an entity of the type “hotel” and, in response, limit a search to only entities that are hotels instead of searching across multiple entities including entities that are not hotels.

Entities in Semantic Dependency Trees

Semantic Dependency Tree

This is an interesting approach to an entity seeking queries. Determining an entity type that may correspond to an entity sought by a query based on a term represented by a root of a dependency tree includes:

Determining the term represented by the root of the dependency tree represents a type of entity.

Determining an entity type that corresponds to an entity sought by the query based on a term represented by a root of the dependency tree includes:

Identifying a node in the tree that represents a term that represents a type of entity
Includes a direct child that represents a term that indicates an action to perform.
In response to determining that the root represents a term that represents and type of entity and includes a direct child that represents a term that indicates an action, identifying the root.

In some implementations, identifying a particular entity based on both the entity type and relevance of the entity to the terms in the query includes:

  • Determining a relevance threshold based on the entity type
  • Determining a relevance score of the particular entity based on the query satisfies the relevance threshold
  • In response to determining the relevance score of the particular entity based on the query satisfies the relevance threshold, identifying the particular entity

This patent on Entity Seeking Queries can be found at:

Answering Entity-Seeking Queries
Inventors: Mugurel Ionut Andreica, Tatsiana Sakhar, Behshad Behzadi, Marcin M. Nowak-Przygodzki, and Adrian-Marius Dumitran
US Patent Application: 20190370326
Published: December 5, 2019
Filed: May 29, 2018

Abstract

In some implementations, a query that includes a sequence of terms is obtained, the query is mapped, based on the sequence of the terms, to a dependency tree that represents dependencies among the terms in the query, an entity type that corresponds to an entity sought by the query is determined based on a term represented by a root of the dependency tree, a particular entity is identified based on both the entity type and relevance of the entity to the terms in the query, and a response to the query is provided based on the particular entity that is identified.

Mapping a Query to a Semantic Dependency Tree

A process that handles entity seeking queries

This process includes:

  • A query mapper that maps a query including a sequence of terms to a semantic dependency tree
  • An entity type identifier that may determine an entity type based on the semantic dependency tree
  • An entity identifier that may receive the query
  • The entity type that is determined
  • Data from various data stores and identify an entity
  • Subquery resolver that may partially resolve the query based on the entity that is identified
  • Query responder that may provide a response to the query

An Example Semantic Dependency Tree

This is how a Semantic Dependency Tree may be constructed:

  1. A semantic dependency tree for a query may be a graph that includes nodes
  2. Each node represents one or more terms in a query
  3. Directed edges originating from a first node and ending at a second node may indicate that the one or more terms represented by the first node are modified by the one or more terms represented by the second node
  4. A node at which an edge ends may be considered a child of a node from which the edge originates
  5. A root of a semantic dependency tree may be a node representing one or more terms that do not modify other terms in a query and are modified by other terms in the query
  6. A semantic dependency tree may only include a single root

An Entity Type Identifier

An entity type identifier may determine an entity type that corresponds to an entity sought by the query based on a term represented by a root of the semantic dependency tree.

For example, the entity type identifier may determine an entity type of “Chinese restaurant” that corresponds to an sought by the query “Call the Chinese restaurant on Piccadilly Street 15” based on the term “Chinese restaurant” represented by the root of the semantic dependency tree.

In another example, the entity type identifier may determine an entity type of “song” for the query “play the theme song from the Titanic” based on the term “play” represented by the root of the semantic dependency tree for the query not representing an entity type and determining that the root has a child that represents the terms “the theme song” which does represent an entity type of “song.”

Entities from a Location History of a Searcher

The entity identifier may extract all the entities from a mobile location history of a searcher which have a type identified by the entity type identifier, such as hotels, restaurants, universities, etc. along with extracting features associated to each such entity such as the time intervals when the user visited the entity or was near the entity, or how often each entity was visited or the user was near the entity.

Entities from a Past Interaction History of a Searcher

In addition to that location history, the entity identifier may extract all the entities that the user was interested in their past interactions that have a type identified by the entity type identifier, such as:

  • Movies that the user watched
  • Songs that the user listened to
  • Restaurants that the user looked up and showed interest in or booked
  • Hotels that the user booked
  • Etc.

Confidence in Relevance for Entity Seeing Queries

The patent also tells us that the entity identify may obtain a relevance score for each entity that reflects a confidence that the entity is sought to be the query.

The relevance score may be determined based on one or more of the features extracted from the data stores that led to the set of entities being identified, the additional features extracted for each entity in the set of entities, and the features extracted from the query.


Copyright © 2020 SEO by the Sea ⚓. This Feed is for personal non-commercial use only. If you are not reading this material in your news aggregator, the site you are looking at may be guilty of copyright infringement. Please contact SEO by the Sea, so we can take appropriate action immediately.
Plugin by Taragana

The post Entity Seeking Queries and Semantic Dependency Trees appeared first on SEO by the Sea ⚓.


SEO by the Sea ⚓


What Clients Want

August 30, 2020 No Comments

An experienced Group Account Director highlights the top requests that clients ask of their agency and how to deliver them.

Read more at PPCHero.com
PPC Hero



The week’s biggest IPO news had nothing to do with Monday’s S-1 deluge

August 30, 2020 No Comments

Welcome back to The TechCrunch Exchange, a weekly startups-and-markets newsletter. It’s broadly based on the daily column that appears on Extra Crunch, but free, and made for your weekend reading. (You can sign up for the newsletter here!)

Ready? Let’s talk money, startups and spicy IPO rumors.

The week’s biggest IPO news had nothing to do with Monday’s S-1 deluge

During Monday’s IPO wave I was surprised to see Asana join the mix. 

After news had broken in June that the company had raised hundreds of millions in convertible debt, I hadn’t guessed that the productivity unicorn wouldn’t give us an S-1 in the very next quarter. I was contentedly wrong. But the reason why Asana’s IPO is notable isn’t really much to do with the company itself, though do take the time to dig into its results and history

What matters about Asana’s debut is that it appears set to test out a model that, until very recently, could have become the new, preferred way of going public amongst tech companies. 

Here’s what I mean: Instead of filing to go public, and raising money in a traditional IPO, or simply listing directly, Asana executed two, large, convertible debt offerings pre-debut, thus allowing it to direct list with lots of cash without having raised endless equity capital while private.

The method looked like a super-cool way to get around the IPO pricing issue that we’ve seen, and also provide a ramp to direct listing for companies that didn’t get showered with billions while private. (That Asana co-founder Dustin Moskovitz’s trust led the debt deal is simply icing on this particular Pop-Tart).

This brief column was going to be all about how we may see unicorns follow the Asana route in time, provided that its debt-powered direct listing goes well. But then the NYSE got permission from the SEC to allow companies to raise capital when they direct-list.

In short, some companies that direct-list in the future will be able to sell a bloc of shares at a market-set value that would have previously set their “open” price. So instead of flogging the stock and setting a price and selling shares to rich folks and then finding out what public investors would really pay, all that IPO faff is gone and bold companies can simply offer shares at whatever price the market will bear. 

All that is great and cool, but as companies will be able to direct-list and raise capital, the NYSE’s nice news means that Asana is blazing a neat trail, but perhaps not one that will be as popular as we had expected.

The NASDAQ is working to get in on the action. As Danny said yesterday on the show, this new NYSE method is going to crush traditional IPOs, provided that we’re understanding it during this, its nascent period.

Market Notes

Look, this week was bananas, and my brain is scrambled toast. You, like myself, are probably a bit confused about how it is only finally Saturday and not the middle of next week. But worry not, I have a quick roundup of the big stuff from our world. And, notes from calls with the COO of Okta and the CEO of Splunk, from after their respective earnings report: 

Over to our chats, starting with Okta COO and co-founder Frederic Kerrest:

  • Okta had a good quarter. But instead of noodling on just the numbers, we wanted to chat with its team about the accelerating digital transformation and what they are seeing in the market. 
  • On the SMB side, Kerrest reported little to no change. This is a bit more bullish than we anticipated, given that it seemed likely that SMB customers would have taken the largest hit from COVID.
  • Kerrest also told us some interesting stuff about how the wave of COVID-related spend has changed: “We actually have seen the COVID ‘go home and remote work very quickly’ [thing], we’ve actually seen that rush subside a little bit, because you know now we’re five months into [the pandemic], so they had to figure it out.”
  • This is a fascinating comment for the startup world
  • Okta is big and public and is going to grow fine for a while. Whatever. For smaller companies aka startups that were seeing COVID-related tailwinds, I wonder how common seeing “that rush subside a little bit” is. If it is very common, many startups that had taken off like a rocket could be seeing their growth come back to Earth.
  • And if they raised a bunch of money off the back of that growth at a killer valuation, they may have just ordered shoes that they’ll struggle to grow into.

And then there was new McLaren F-1 sponsor Splunk, data folks who are in the midst of a transition to SaaS that is seeing the firm double-down on building ARR and letting go of legacy incomes:

  • I spoke with CEO Doug Merritt, kicking off with a question about his use of the word “tectonic” regarding the shift to data-driven decisions from Splunk’s earnings report. (“As organizations continue to adapt to tectonic societal shifts brought on by COVID-19, one thing is constant: the power of data to radically transform business.”)
  • I wanted to know how far down the American corporate stack that idea went; are mid-size businesses getting more data-savvy? What about SMBs? Merritt was pretty bullish: “We’re getting to tectonic,” he said during our call, adding that before “it really was the Facebooks, the Googles, the Apples, the DoorDashes, [and] the LinkedIns that were using [Splunk].” But now, he said, even small restaurant chains are using data to better track their performance. 
  • Relating this back to the startup world, I’ve been curious if lots of stuff that you and I think is cool, like low-code business app development, will actually find as wide a footing in the market as some expect. Why? Because most small and medium-sized businesses are not tech companies at all. But if Merritt is right, then the CEO of Appian might be right as well about how many business apps the average company is going to have in a few years’ time.

And finally for Market Notes, my work BFF and IRL friend Ron Miller wrote about Box’s earnings this week, and how the changing world is bolstering the company. It’s worth a read. (Most public software companies are doing well, mind.)

Various and Sundry

We’re already over length, so I’ll have to keep our bits-and-bobs section brief. Thus, only the brightest of baubles for you, my friend:

And with that, we are out of room. Hugs, fist bumps and good vibes, 

Alex


Startups – TechCrunch


Top Tips To Get The Most From Microsoft’s New Shutterstock API

August 29, 2020 No Comments

Microsoft Advertising has given its users access to high-quality images with their new Shuttershock partnership. Here’s how you can make the most of it.

Read more at PPCHero.com
PPC Hero


App store optimization success: Top five KPIs you must measure

August 29, 2020 No Comments

30-second summary:

  • Mobile applications have become essential for human life, It has managed to reach every corner of the world, it is used for all kinds of things.
  • There are more than two million mobile apps available in app stores but all of them are not equally successful.
  • App Store Optimization (ASO) is one of the techniques that app owners and developers used to optimize their mobile for keywords.
  • After that, you need to measure campaign success by monitoring some of the KPI.
  • Today we will discuss what KPI you should keep monitor to measure the success of your app optimization campaign.

Across two major app marketplaces, there are already more than two million mobile apps and still counting. Mobile apps have made inroads across all nooks and corners of our daily lives and enterprise operations. But despite all these, not all mobile apps are equally successful, and there is a multitude of apps that struggle to survive as businesses. Perhaps, App Store Optimization (ASO) is something that app creators must consider.

Naturally, app publishers and marketers worldwide consider it extremely important to monitor several metrics concerning user engagement, user retention, and business conversion. Here we will explain the five leading metrics that you need to track for measuring the success of your app in app stores.

1. Discoverability in App Stores

The most important thing for any app to get traction in app marketplaces is to become easily discoverable and visible to the target audience. Whether coming into search results or getting featured or hitting top charts, in one way or the other, your app needs to be discoverable and visible to the audience. This is also the principal objective of App Store Optimization (ASO).

Some metrics to keep track of discoverability of the app include the following.

Keywords ranking

Where in the search result, your app appears against a target keyword refers to your search ranking.

Top charts ranking

The ranking of your app in various top charts by categories or by parameters such as Free, Premium, and others.

Category ranking

Where your app in the respective category ranks refers to the category ranking.

Featured

Whether the app is listed among the top apps featured by the App Store or Play Store.

All these metrics that can easily be tracked can reveal your app’s discoverability in the app marketplaces.

2. Active users

This is one of the most important metrics to measure the traction and engagement of a mobile app. The number of active users for a mobile app directly shows the audience engagement and how it evolves. When this metric shows growth, that means the app is getting more traction. The active user metrics can further be categorized into four metrics as per audience engagement in different time spans.

Daily Active Users (DAU)

This metric refers to the number of users using the app on a particular day.

Average Daily Active Users (ADAU)

The number of users using the app in a single month is divided by the number of days in a month.

Weekly Active Users (WAU)

The number of users using the app in a single week.

Monthly Active Users (MAU)

The number of users using the app in a particular month.

3. Lifetime Value (LTV)

Whether calculated on a monthly, daily, or weekly basis, the number of users hardly gives an idea of the business conversion or the kind of revenue they generate for the app. The Lifetime Value or LTV is the metric that helps measure the gross revenue generated by a user over a period of time. This metric is more closely related to the bottom line of an app and hence is very important.

Though you can easily track user session time on a weekly, daily, or monthly basis and track their CTR and impressions, measuring the business conversion remains difficult. This is where this metric comes as handy as it allows evaluating the total sum outcome of the entire app marketing efforts.

4. User acquisition cost

Another important metric closely related to the bottom line and revenue is the user acquisition cost. Your total marketing budget spent on user acquisition can be divided by the total number of users to get an idea of the cost of acquisition. The metric further can be divided on a monthly and yearly basis and can be seen whether the cost of acquisition is growing or decreasing.

5. Conversion rate

Your app is easily visible and discoverable as per the various visibility metrics we have discussed. Now the question is, does this visibility convert into app downloads. After discovering your app downloads the app, how many of the visitors is a crucial metric to measure the success of your App Store Optimization and app marketing?

Some of the key methods to boost App Store Optimization and conversion include using great app title, engaging app description powered by screenshots, images and video content, app reviews, and app ratings.

App market conversion can be tracked through the important measurement metric called Click-through rate (CTR). The proportion of people who, after landing on your app marketplace snippet clicks to go into the product page, is expressed in percentage, and it is called the conversion rate. It is an unmistakable part of the conversion funnel that app marketers need to monitor on a regular basis.

To improve the conversion rate, fortunately, you have an array of sophisticated A/B testing tools that helps you to evaluate the impact of various aspects of your app listing and accordingly fees helpful suggestions to improve conversion rate.

Conclusion

While these metrics are already well known and are regularly tracked by the app marketers around the world, you need to make sure to use a good analytics engine to track your audience engagement and business conversion more accurately.

Juned Ghanchi is Co-founder of IndianAppDevelopers, a mobile app development company builds iOS and Android mobile apps for startups to big brands. Juned has over a decade of experience across Software consulting, App solutions, and App development.

The post App store optimization success: Top five KPIs you must measure appeared first on Search Engine Watch.

Search Engine Watch


Entity Seeking Queries and Semantic Dependency Trees

August 29, 2020 No Comments

Entity Seeking Queries and Semantic Dependency Trees

Queries for some searches may be entity seeking queries.

Someone may ask, “What is the hotel that looks like a sail.” That query may look for an entity such as the building, the Burj Al Arab Jumeirah.

That entity may be identified by looking at Semantic Dependency Trees, which answer a question in the query (example below)

Other queries are not entity seeking queries and do not look for answers about specific entities. An example is “What is the weather today?” An answer to a query like that could be, “The weather will be between 60-70 degrees Fahrenheit, and sunny today.”

Actions May Accompany Queries that Seek Entities

Google was granted a patent on answering entity seeking queries.

The process in the patent may perform particular actions for queries seeking one or more entities.

Those actions may include:

  • Identifying one or more entities a query is seeking
  • Determining whether the query seeks one specific entity or more than one

For example, the patent may decide a query “What is the hotel that looks like a sail” is looking for a single entity that is a hotel.

In another example, a query “What restaurants nearby serve omelets” looks for many entities that are restaurants.

Alternatively, the system may perform may find a most relevant entity or entities, and present what is identified to a searcher if it is sufficiently relevant to the query. For example, it may identify that the Burj Al Arab Jumeirah is a hotel and is relevant to the term “looks like a sail,” and, in response, audibly output synthesized speech of “Burj Al Arab Jumeirah.”

Additional Dialog about a Query to Concatenate an Entity Seeking Query

Yet another addition or alternative action may include initiating a dialog with the user for more details about the entities that are sought.

For example, the system may determine that a query is seeking a restaurant and there may be two entities that are restaurants are very relevant to the terms in the query and, in response, ask the searcher “Can you give me more details” and concatenate additional input from the user to the original query and re-execute the concatenated query.

Identifying SubQueries of Entity Seeking Queries

Another additional or alternative action may include identifying subqueries of a query which are entity-seeking, and using the above actions to answer the subquery, and then replacing the subqueries by their answers in the original query to obtain a partially resolved query which can be executed.

For example, the system may receive a query of “Call the hotel that looks like a sail,” determine that “the hotel that looks like a sail” is a subquery that seeks an entity, determine an answer to the subquery is “Burj Al Arab Jumeirah,” in response replace “the hotel that looks like a sail” in the query with “The Burj Al Arab Jumeirah” to obtain a partially resolved query of “Call the Burj Al Arab Jumeirah,” and then executes the partially resolved query.

Looking at Previous Queries

Another additional or alternative action may include identifying that a user is seeking entities and adapting how the system resolve queries accordingly.

For example, the system may determine that sixty percent of the previous five queries that a user searched for in the past two minutes sought entities and, in response, determine that a next query that a user provides is more likely an entity seeking query, and process the query accordingly.

An Advantage From Following this Process

An advantage may be more quickly resolving queries in a manner that satisfies a searcher.

For example, the system may be able to immediately provide an actual answer of “The Burj Al Arab Jumeirah” for the query “What hotel looks like a sail” where another system may instead provide a response of “no results found” or provide a response that is a search result listing for the query.

Entity Seeking Queries and Semantic Dependency Trees

Entity Seeking Queries
Another advantage may be that the process may be able to more efficiently identify an entity sought by a query. For example, it may determine an entity seeking query is looking for an entity of the type “hotel” and, in response, limit a search to only entities that are hotels instead of searching across multiple entities including entities that are not hotels.

Entities in Semantic Dependency Trees

Semantic Dependency Tree

This is an interesting approach to an entity seeking queries. Determining an entity type that may correspond to an entity sought by a query based on a term represented by a root of a dependency tree includes:

Determining the term represented by the root of the dependency tree represents a type of entity.

Determining an entity type that corresponds to an entity sought by the query based on a term represented by a root of the dependency tree includes:

Identifying a node in the tree that represents a term that represents a type of entity
Includes a direct child that represents a term that indicates an action to perform.
In response to determining that the root represents a term that represents and type of entity and includes a direct child that represents a term that indicates an action, identifying the root.

In some implementations, identifying a particular entity based on both the entity type and relevance of the entity to the terms in the query includes:

  • Determining a relevance threshold based on the entity type
  • Determining a relevance score of the particular entity based on the query satisfies the relevance threshold
  • In response to determining the relevance score of the particular entity based on the query satisfies the relevance threshold, identifying the particular entity

This patent on Entity Seeking Queries can be found at:

Answering Entity-Seeking Queries
Inventors: Mugurel Ionut Andreica, Tatsiana Sakhar, Behshad Behzadi, Marcin M. Nowak-Przygodzki, and Adrian-Marius Dumitran
US Patent Application: 20190370326
Published: December 5, 2019
Filed: May 29, 2018

Abstract

In some implementations, a query that includes a sequence of terms is obtained, the query is mapped, based on the sequence of the terms, to a dependency tree that represents dependencies among the terms in the query, an entity type that corresponds to an entity sought by the query is determined based on a term represented by a root of the dependency tree, a particular entity is identified based on both the entity type and relevance of the entity to the terms in the query, and a response to the query is provided based on the particular entity that is identified.

Mapping a Query to a Semantic Dependency Tree

A process that handles entity seeking queries

This process includes:

  • A query mapper that maps a query including a sequence of terms to a semantic dependency tree
  • An entity type identifier that may determine an entity type based on the semantic dependency tree
  • An entity identifier that may receive the query
  • The entity type that is determined
  • Data from various data stores and identify an entity
  • Subquery resolver that may partially resolve the query based on the entity that is identified
  • Query responder that may provide a response to the query

An Example Semantic Dependency Tree

This is how a Semantic Dependency Tree may be constructed:

  1. A semantic dependency tree for a query may be a graph that includes nodes
  2. Each node represents one or more terms in a query
  3. Directed edges originating from a first node and ending at a second node may indicate that the one or more terms represented by the first node are modified by the one or more terms represented by the second node
  4. A node at which an edge ends may be considered a child of a node from which the edge originates
  5. A root of a semantic dependency tree may be a node representing one or more terms that do not modify other terms in a query and are modified by other terms in the query
  6. A semantic dependency tree may only include a single root

An Entity Type Identifier

An entity type identifier may determine an entity type that corresponds to an entity sought by the query based on a term represented by a root of the semantic dependency tree.

For example, the entity type identifier may determine an entity type of “Chinese restaurant” that corresponds to an sought by the query “Call the Chinese restaurant on Piccadilly Street 15” based on the term “Chinese restaurant” represented by the root of the semantic dependency tree.

In another example, the entity type identifier may determine an entity type of “song” for the query “play the theme song from the Titanic” based on the term “play” represented by the root of the semantic dependency tree for the query not representing an entity type and determining that the root has a child that represents the terms “the theme song” which does represent an entity type of “song.”

Entities from a Location History of a Searcher

The entity identifier may extract all the entities from a mobile location history of a searcher which have a type identified by the entity type identifier, such as hotels, restaurants, universities, etc. along with extracting features associated to each such entity such as the time intervals when the user visited the entity or was near the entity, or how often each entity was visited or the user was near the entity.

Entities from a Past Interaction History of a Searcher

In addition to that location history, the entity identifier may extract all the entities that the user was interested in their past interactions that have a type identified by the entity type identifier, such as:

  • Movies that the user watched
  • Songs that the user listened to
  • Restaurants that the user looked up and showed interest in or booked
  • Hotels that the user booked
  • Etc.

Confidence in Relevance for Entity Seeing Queries

The patent also tells us that the entity identify may obtain a relevance score for each entity that reflects a confidence that the entity is sought to be the query.

The relevance score may be determined based on one or more of the features extracted from the data stores that led to the set of entities being identified, the additional features extracted for each entity in the set of entities, and the features extracted from the query.


Copyright © 2020 SEO by the Sea ⚓. This Feed is for personal non-commercial use only. If you are not reading this material in your news aggregator, the site you are looking at may be guilty of copyright infringement. Please contact SEO by the Sea, so we can take appropriate action immediately.
Plugin by Taragana

The post Entity Seeking Queries and Semantic Dependency Trees appeared first on SEO by the Sea ⚓.


SEO by the Sea ⚓


TikTok’s rivals in India struggle to cash in on its ban

August 28, 2020 No Comments

For years, India has served as the largest open battleground for Silicon Valley and Chinese firms searching for their next billion users.

With more than 400 million WhatsApp users, India is already the largest market for the Facebook-owned service. The social juggernaut’s big blue app also reaches more than 300 million users in the country.

Google is estimated to reach just as many users in India, with YouTube closely rivaling WhatsApp for the most popular smartphone app in the country.

Several major giants from China, like Alibaba and Tencent (which a decade ago shut doors for most foreign firms), also count India as their largest overseas market. At its peak, Alibaba’s UC Web gave Google’s Chrome a run for its money. And then there is TikTok, which also identified India as its biggest market outside of China.

Though the aggressive arrival of foreign firms in India helped accelerate the growth of the local ecosystem, their capital and expertise also created a level of competition that made it too challenging for most Indian firms to claim a slice of their home market.

New Delhi’s ban on 59 Chinese apps on June 30 on the basis of cybersecurity concerns has changed a lot of this.

Indian apps that rarely made an appearance in the top 20 have now flooded the charts. But are these skyrocketing download figures translating to sustaining users?

An industry executive leaked the download, monthly active users, weekly active users and daily active users figures from one of the top mobile insight firms. In this Extra Crunch report, we take a look at the changes New Delhi’s ban has enacted on the world’s second largest smartphone market.

TikTok copycats

Scores of startups in India, including news aggregator DailyHunt, on-demand video streamer MX Player and advertising giant InMobi Group, have launched their short-video format apps in recent months.


Social – TechCrunch


TikTok’s rivals in India struggle to cash in on its ban

August 28, 2020 No Comments

For years, India has served as the largest open battleground for Silicon Valley and Chinese firms searching for their next billion users.

With more than 400 million WhatsApp users, India is already the largest market for the Facebook-owned service. The social juggernaut’s big blue app also reaches more than 300 million users in the country.

Google is estimated to reach just as many users in India, with YouTube closely rivaling WhatsApp for the most popular smartphone app in the country.

Several major giants from China, like Alibaba and Tencent (which a decade ago shut doors for most foreign firms), also count India as their largest overseas market. At its peak, Alibaba’s UC Web gave Google’s Chrome a run for its money. And then there is TikTok, which also identified India as its biggest market outside of China.

Though the aggressive arrival of foreign firms in India helped accelerate the growth of the local ecosystem, their capital and expertise also created a level of competition that made it too challenging for most Indian firms to claim a slice of their home market.

New Delhi’s ban on 59 Chinese apps on June 30 on the basis of cybersecurity concerns has changed a lot of this.

Indian apps that rarely made an appearance in the top 20 have now flooded the charts. But are these skyrocketing download figures translating to sustaining users?

An industry executive leaked the download, monthly active users, weekly active users and daily active users figures from one of the top mobile insight firms. In this Extra Crunch report, we take a look at the changes New Delhi’s ban has enacted on the world’s second largest smartphone market.

TikTok copycats

Scores of startups in India, including news aggregator DailyHunt, on-demand video streamer MX Player and advertising giant InMobi Group, have launched their short-video format apps in recent months.


Startups – TechCrunch


Instagram Guides may soon allow creators to recommended places, products and more

August 27, 2020 No Comments

Instagram is working to expand its recently launched “Guides” feature which initially debuted with a specific focus on wellness content. The feature, which launched in May, has allowed select organizations and experts to share resources related to managing your mental health — including things like handling anxiety or grief amid the COVID-19 pandemic, for example. A handful of creators first gained access to the feature, and have since posted their wellness tips on their Instagram profiles in a separate tab, called “Guides.” Now, Instagram is developing tools that will allow creators to build out Guides for other types of tips and recommendations, too — like recommended places or even recommended products.

The larger goal with Guides is to give Instagram users a way to post longer-form content that’s not just a photo or video. Currently, Guides can include photos, galleries and videos sourced from either the creator’s own profile, which is more common, or from other creators. In addition, the Guides include commentary or tips alongside the media.

Instagram Guides

Instagram Guides today (Image Credits: Instagram)

The feature would allow creators to use Instagram as their platform for sharing tips and advice, instead of having that traffic redirected outside of Instagram — like to a blog or other website.

At launch, Instagram head Adam Mosseri said the Guides feature was originally designed with the travel use case in mind, but the company pivoted Guides to focus on wellness because of the COVID-19 pandemic.

Now it appears Instagram will be returning to its original idea of letting creators build Guides for places — and for other things, too.

The changes to Guides were first uncovered by Twitter user and self-described leaker, Alessandro Paluzzi. He tells TechCrunch he found the new features by reverse engineering the Instagram app. But these changes haven’t yet launched to the wider Instagram user base.

Instagram tests new feature

Image Credits: Alessandro Paluzzi, via Twitter

The tests show the company experimenting with a new compose screen, as well. Here, users are presented with all the different ways you can publish to Instagram’s social network. This includes the option to create a new Feed Post, post a Story or Story Highlight, post to IGTV, post to Reels or create a new Guide.

If you choose “Guide” from the list, you’re then presented with a menu that asks you to choose a Guide type. This can be a Places Guide, for recommending favorite places; a Products Guide, for recommending favorite products; or a Posts Guide, which is a more general-purpose format for recommending a series of your favorite Instagram posts.

This feature would allow Guides to easily fit into Instagram influencers’ workflows, as they often make recommendations to followers about where to go, what to purchase and more. Creators could even increase their affiliate network revenue or direct more users to their sponsored posts through the use of Guides, if they chose.

Instagram tests new feature

Image Credits: Alessandro Paluzzi, via Twitter

Instagram confirmed the new features are part of a series of improvements to Guides it’s working on.

“This is part of an early test as we work to improve guides. We’ll have more to share soon,” a spokesperson said. The company declined to say if or when the changes would roll out more broadly, adding it’s still in the early stages and the product could change based on user feedback. Instagram also declined to speak to its long-term plans for the Guides feature.

The changes come shortly after Pinterest began edging its way into Instagram territory. The social pinboarding site recently began testing its own new feature aimed at aggregating content for longer-form storytelling. With Story Pins, Pinterest creators could build out “guides” of their own for topics like recipes, crafts, DIY projects or more. In addition, more users are turning to Facebook rival TikTok for tips, inspiration and other creator content.


Social – TechCrunch